Advances in Operational Near Real-Time SMAP Soil Moisture Retrieval Processing

Soil moisture is a crucial data point for weather prediction and climate projection systems, supplying vital details for the management of floods and droughts, in addition to planning irrigation schedules in agriculture. In this research, we aim to produce near real-time soil moisture at a spatial resolution of 9-km by incorporating an Operational Enhanced Soil Moisture Active Passive retrieval processor, henceforth referred as SMAP_E_OPL, within the NASA's Land surface Data Toolkit (LDT). The single channel retrieval method used for SMAP Level 2 Enhanced soil moisture products (i.e., SMAP_L2SMP_E) is used here by SMAP_E_OPL processor. However, the SMAP_E_OPL uses an Inverse Distance Squared Weighted (IDSW) interpolation approach to resample the 36-km resolution SMAP L1B brightness temperature (TB) data to 9-km in contrast to SMAP_L2SMP_E which uses the Backus-Gilbert (BG) optimal interpolation method. Using IDSW resampling method reduces the computational complexity of soil moisture retrievals up to 3 minutes per half-orbit compared to BG interpolation method. Assessing the soil moisture retrievals by SMAP_E_OPL, conducted via comparing them with in situ soil moisture measurements throughout the continental United States, suggests that SMAP_E_OPL produces near real-time soil moisture at a 9 km resolution with an accuracy level similar to the official (high latency) SMAP enhanced soil moisture products. The difference between SMAP_E_OPL and SMAP_L2SMP_E is less than 0.07 m3 m-3 in Root Mean Square Difference (RMSD) for 90% of the land grids over the global domain. This result indicate the feasibility of the SMAP_E_OPL processor to provide the near real-time high spatial resolution soil moisture data for operational forecasting, data assimilation, and monitoring systems.