Integration of microwave data from SMAP and AMSR2 for soil moisture monitoring in Italy

In this study, an integration of microwave data obtained from the SMAP and AMSR2 satellite radiometers has been attempted, to achieve an accurate estimation of the Soil Moisture Content (SMC). This research aimed to overcome the failure of radar sensor in SMAP satellite as well as the failure to generate the radar/radiometer combined SMC product at a spatial resolution of 9 km×9 km. A disaggregation technique, based on the Smoothing Filter based Intensity Modulation (SFIM), enabled us to obtain co-located SMAP and AMSR2 brightness measurements at L, C, X, Ku and Ka bands at approximately 10 km×10 km on the selected test area, which corresponds to the entire Italian territory. These disaggregated microwave data were used as inputs of the "HydroAlgo" retrieval algorithm based on Artificial Neural Networks (ANN), which were able to exploit the synergy between radiometric acquisitions from these two sensors. The algorithm was defined, implemented and tested using all the overlapping orbits of SMAP and AMSR2 over Italy throughout the 9_month period between April and December 2015. Distributed SMC reference values for implementing and validating the algorithm were obtained from the Soil Water Balance hydrological model, SWBM. Through HydroAlgo, an SMC product at a resolution of approximately 10 km×10 km was obtained. This result is close to the original Radar/Radiometer SMC product from SMAP, with an average correlation coefficient R > 0.75 and RMSE ? 0.03m3/m3, in both ascending and descending orbits.In this study, an integration of microwave data obtained from the SMAP and AMSR2 satellite radiometers has been attempted, to achieve an accurate estimation of the Soil Moisture Content (SMC). This research aimed to overcome the failure of radar sensor in SMAP satellite as well as the failure to generate the radar/radiometer combined SMC product at a spatial resolution of 9 km×9 km. A disaggregation technique, based on the Smoothing Filter based Intensity Modulation (SFIM), enabled us to obtain co-located SMAP and AMSR2 brightness measurements at L, C, X, Ku and Ka bands at approximately 10 km×10 km on the selected test area, which corresponds to the entire Italian territory. These disaggregated microwave data were used as inputs of the "HydroAlgo" retrieval algorithm based on Artificial Neural Networks (ANN), which were able to exploit the synergy between radiometric acquisitions from these two sensors. The algorithm was defined, implemented and tested using all the overlapping orbits of SMAP and AMSR2 over Italy throughout the 9_month period between April and December 2015. Distributed SMC reference values for implementing and validating the algorithm were obtained from the Soil Water Balance hydrological model, SWBM. Through HydroAlgo, an SMC product at a resolution of approximately 10 km×10 km was obtained. This result is close to the original Radar/Radiometer SMC product from SMAP, with an average correlation coefficient R > 0.75 and RMSE ? 0.03m3/m3, in both ascending and descending orbits.