|
Observations from spaceborne microwave (MW) and infrared (IR) passive sensors are the backbone of current satellite meteorology, essential for data assimilation into modern numerical weather prediction and climate benchmarking. In this context, over the last decades, the study and the analysis of cloud microphysics have received increasing attention to better understand cloud feedbacks on climate. MW and IR observations from space offer complementary features concerning cloud microphysics, and various tools have been developed to retrieve cloud parameters such as the effective radius of water and ice clouds. However, MW-IR synergy for cloud investigation is currently under-explored. In this framework, innovative processing methods, such as those based on the use of Artificial Intelligence (AI), which can run on large databases and can handle hundreds of input variables from different sensors, such as those operating in hyperspectral and multispectral channels of the infrared and the microwave bands, such as the New Generation Atmospheric Sounding Interferometer (IASI-NG) and the Microwave Sounder (MWS) of the EPS second generation (EPSSG) platforms whose forthcoming launch is scheduled from 2024 onwards. A regression framework has been implemented based on the combined use of Random Forest (RF) regression and the principal components analysis (PCA) of IASI-NG and MWS observations to input the RF regressors. The supervised learning of liquid and ice water clouds' effective radii was carried out based on this framework. In conclusion, the regression analysis shows good agreement between reference and retrieved effective radius, with 80% correlation and root-mean-square error (RMSE) of 0.68 μm for liquid and 11.6 μm for ice cloud effective radius.
|
