Rainrate Estimation from FY-4A Cloud Top Temperature for Mesoscale Convective Systems by Using Machine Learning Algorithm

Satellite rainrate estimation is a great challenge, especially in mesoscale convective systems (MCSs), which mainly due to the absence of direct physical connection between observable cloud parameters and surface rainrate. The machine learning technique was employed this study estimate MCS domain via using top temperature (CTT) derived from geostationary satellite. Five kinds models were investigated, i.e., polynomial regression, support vector machine, decision tree, random forest, multilayer perceptron, precipitation Climate Prediction Center morphing (CMORPH) used as reference. A total 31 CTT related features designed be potential inputs for training an algorithm, they all proved have positive contribution modulating algorithm. Random forest (RF) shows best performance among five models. By combining classification regression schemes RF model, RF-based hybrid algorithm proposed first discriminate rainy pixel then its For samples considered study, such generates estimation, accuracy definitely higher than operational product FY-4A. These results demonstrate promising feasibility applying solve satellite retrieval problem.