Remote-Sensing-Based Water Balance for Monitoring of Evapotranspiration and Water Stress of a Mediterranean Oak–Grass Savanna

Abstract

Mediterranean oak savannas (known as dehesas in Spain) are exposed to numerous threats from natural and economic causes. A close monitoring of the use of water resources and the status of the vegetation in these ecosystems can be useful tools for maintaining the production of ecological services. This study explores the estimation of evapotranspiration (ET) and water stress over a dehesa by integrating remotely sensed data into a water balance using the FAO-56 approach (VI-ETo model). Special attention is paid to the different phenology and contribution to the system's hydrology of the two main canopy layers of the system (tree + grass). The results showed that the model accurately reproduced the dynamics of the water consumed by the vegetation, with RMSE of 0.47 mm day(-1) and low biases for both, the whole system and the grass layer, when compared with flux tower measurements. The ET/ETo ratio helped to identify periods of water stress, confirmed for the grassland by measured soil water content. The modeling scheme and Sentinel-2 temporal resolution allowed the reproduction of fast and isolated ET pulses, important for understanding the hydrologic behavior of the system, confirming the adequacy of this sensor for monitoring grasslands water dynamics.