27-Year Climatology of Severe Wind-Producing Mesoscale Convective Systems in the United States

Mesoscale convective systems (MCSs) account for a large fraction of widespread damaging wind events handled by the National Weather Service’s Storm Prediction Center (SPC). SPC seeks an operationally oriented categorization of MCSs by severe wind production, up to and including derechos (Hinrichs 1888; Johns and Hirt 1987). While recent climatologies have successfully identified MCSs with satellite data or via machine learning classification of radar features, this operational application requires articulable radar-based criteria for low-level MCS organization, as well as matching of local storm reports to these MCSs. We define MCS objects in radar reflectivity mosaics by excluding non-convective or weakly convective regions, merging convective elements very close to one another, and searching for convective lines of at least 100 km. To qualify as an MCS, a 100-km convective line must be tracked at 15-minute intervals for at least 3 h, length and time scales consistently applied to MCSs in the literature (e.g., Parker and Johnson 2000; Haberlie and Ashley 2019). Preliminarily, this automated algorithm has identified over 21,000 MCSs over the contiguous United States from 1996 through 2022. We show the spatial, seasonal, and diurnal distributions of MCS occurrence, as well as synoptic to mesoscale environmental composites for key subgroups of MCSs. Matching of local storm reports to each MCS at each 15-minute time step enables classification by severity. A companion presentation (Squitieri et al. 2023) previews such a classification system, including a proposed formal objective definition of a derecho that draws on the varying definitions in past studies.