Combining land surface models and remote sensing data to estimate evapotranspiration for drought monitoring in Europe

The main hydrologic feedback from the land-surface to the atmosphere is the evapotranspiration, ET, which embraces the response of both the soil and vegetated surface to the atmospheric forcing (e.g., precipitation and temperature), as well as influences locally atmospheric humidity, cloud formation and precipitation, the main driver for drought. Actual ET is regulated by several factors, including biological quantities (e.g., rooting depth, leaf area, fraction of absorbed photosynthetically active radiation) and soil water status. The ET temporal dynamic is strongly affected by rainfall deficits, and in turn it represents a robust proxy of the effects of water shortage on plants. These characteristics make ET a promising quantity for monitoring environmental drought, defined as a shortage of water availability that reduces the ecosystem productivity. In the last few decades, the capability to accurately model ET over large areas in a spatial-distributed fashion has increased notably. Most of the improvements in this field are related to the increasing availability of remote sensing data, and the achievements in modelling of ET-related quantities. Several land-surface models exploit the richness of newly available datasets, including the Community Land Model (CLM) and the Meteosat Second Generation (MSG) ET outputs. Here, the potentiality of ET maps obtained by combining land-surface models and remote sensing data through these two schemes is explored, with a special focus on the reliability of ET (and derived standardized variables) as drought indicator. Tests were performed over Europe at moderate spatial resolution (3-5 km), with the final goal to improve the estimation of soil water status as a contribution to the European Drought Observatory (EDO, http://edo.jrc.ec.europa.eu).