A Climatology of Quasi-linear Convective Systems in the U.S.

This study determines the degree that quasi-linear convective systems (QLCSs) threaten humans and their assets by: 1) generating a climatology of QLCSs; and 2) determining the proportion of QLCSs that produce severe weather. U.S. composite reflectivity mosaics and machine learning techniques are used to identify QLCSs. The machine learning model is trained and validated using spatial and intensity information from thousands of manually-labelled QLCS and Non-QLCS events. Convective regions determined by the model to be QLCSs are used as geographic foci for spatiotemporal filtering of severe reports. This work discusses the utility of this approach for automated storm-mode classification in the context of severe weather occurrence, as well as the climatological results corresponding to the occurrence of severe and sub-severe QLCSs.