Filling the gap: Using severe storm climatologies to investigate the predictability and dynamics of precipitation episodes in the warm season

An important subset of the research focused on improving forecasting performance during the warm season (May – August) involves forecasting severe weather – tornadoes in particular – but also large hail, and strong and damaging winds. One promising avenue for improved probabilistic prediction of warm season precipitation lies in the potential to exploit intrinsic predictability associated with coherent events of long-lived precipitation, as described by Carbone et al. (JAS, 2002) and others. We adopt the methodology of Carbone et al., but apply these techniques to the severe weather databases compiled by the Storm Prediction Center (SPC) in order to determine how well the implications of intrinsic predictability carry over to the problem of predicting severe weather events.

Comparisons were made between the radar data used by Carbone et al. and others to diagnose the existence of coherent precipitation episodes in the warm season in the central United States and the SPC databases of severe weather events, including tornadoes, significant hail, and high and damaging winds. This approach has yielded insights into both the radar and severe weather datasets and the relationships between them. Although there was some similarity, the differences among the datasets were often more apparent. The diurnal cycles of severe weather occurrence, in particular, show significant deviations from the diurnal cycle of convective precipitation described by Carbone et al. and others. We will discuss the implications of our findings in the context of current understanding of mesoscale dynamics.