Soil Moisture Active Passive Improves Global Soil Moisture Simulation in a Land Surface Scheme and Reveals Strong Irrigation Signals Over Farmlands

Abstract The successful Soil Moisture Active Passive (SMAP) mission provides operational soil moisture products of high quality; yet its impacts on global carbon and water cycle estimation are yet to be further investigated. Here we assimilated the SMAP enhanced Level‐2 soil moisture product at 9 km resolution into a land surface scheme in order to study the soil moisture control on the functioning of terrestrial ecosystems. We found that SMAP significantly improves soil moisture simulations, especially in the spring. Extensive wetting signals were revealed over croplands in arid and semi‐arid regions and could not be explained using reanalysis meteorological data, indicating an additional water input, for example, irrigation. Stronger impacts on gross primary production and evapotranspiration simulations are found in wetting adjustments than in drying adjustments after data assimilation. This study suggests that the performance of the land surface scheme benefits greatly from assimilating the SMAP soil moisture product. Plain Language Summary Soil Moisture Active Passive (SMAP) is a satellite that can measure the moisture content of soil surface. The soil moisture map from this satellite is expected to help us to track drought and flood events and to understand how plants respond to water conditions. In this study, we demonstrate whether a new SMAP soil moisture product at finer spatial resolution (9 km) is helpful for better simulations of soil moisture, water use and vegetation growth using a computer model. We found that SMAP significantly improves soil moisture simulations, especially in the spring. Thanks to this enhanced SMAP product, we also found that it reveals strong irrigation signals over farmlands in arid and semi‐arid areas, while many of these irrigation signals are impossible to detect from the standard SMAP product at a coarser resolution (36 km). Compared to previous studies which mostly focused on small regions, our study demonstrated the usefulness of SMAP at the global scale for agricultural applications. Key Points Assimilation of the Soil Moisture Active Passive (SMAP) product at 9 km resolution in an ecosystem model improves soil moisture simulation Strong and clustered irrigation signals are revealed over farmlands in arid and semi‐arid areas Non‐symmetric impacts on gross primary production and evapotranspiration simulations between wetting and drying adjustments are found