Stratiform and Convective Rain Classification Using Machine Learning Models and Micro Rain Radar

Rain type classification into convective and stratiform is an essential step required to improve quantitative precipitation estimations by remote sensing instruments. Previous studies with Micro Radar (MRR) measurements subjective rules have been performed classify rain events. However, automating this process using machine learning (ML) models provides the advantages of fast reliable possibility minute minute. A total 20,979 min data measured MRR at Das in northeast Spain were used build seven types ML for classification. The proposed use a set 22 parameters that summarize reflectivity, Doppler velocity, spectral width (SW) above below so-called separation level (SL). This defined as highest increase velocity corresponds bright band rain. pre-classification each based on microstructure provided collocated disdrometer was performed. Our results indicate complex models, particularly tree-based ensembles such xgboost random forest which capture interactions different features, perform better than simpler models. Applying methods from field interpretable ML, we identified reflectivity lowest layer average layers SL most important features. High low SW values higher probability