Integrating process-related information into an artificial neural network for root-zone soil moisture prediction

Quantification of root-zone soil moisture (RZSM) is crucial for agricultural applications and the soil sciences. RZSM impacts processes such as vegetation transpiration and water percolation. Surface soil moisture (SSM) can be assessed through active and passive microwave remotesensing methods, but no current sensor enables direct RZSM retrieval. Spatial maps of RZSM can be retrieved via proxy observations (vegetation stress, water storage change and surface soil moisture) or via land surface model predictions. In this study, we investigated the combination of surface soil moisture information with process-related inferred features involving artificial neural networks (ANNs). We considered the infiltration process through the soil water index (SWI) computed with a recursive exponential filter and the evaporation process through the evaporation efficiency computed based on a Moderate Resolution Imaging Spectroradiometer (MODIS) remote-sensing dataset and a simplified analytical model, while vegetation growth was not modeled and was only inferred through normalized difference vegetation index (NDVI) time series. Several ANN models with different sets of features were developed. Training was conducted considering in situ stations distributed in several areas worldwide characterized by different soil and climate patterns of the International Soil Moisture Network (ISMN), and testing was applied to stations of the same data-hosting facility. The results indicate that the integration of process-related features into ANN models increased the overall performance over the reference model level in which only SSM features were con sidered. In arid and semiarid areas, for instance, performance enhancement was observed when the evaporation efficiency was integrated into the ANN models. To assess the robustness of the approach, the trained models were applied to observation sites in Tunisia, Italy and southern India that are not part of the ISMN. The results reveal that joint use of surface soil moisture, evaporation efficiency, NDVI and recursive exponential filter represented the best alternative for more accurate predictions in the case of Tunisia, where the mean correlation of the predicted RZSM based on SSM only sharply increased from 0.443 to 0.801 when process-related features were integrated into the ANN models in addition to SSM. However, process-related features have no to little added value in temperate to tropical conditions.