Developmental Work at the Storm Prediction Center in Pursuit of Tornadic Supercell Probability and Tornado Intensity Estimation Using a Severe Supercell Dataset

The Storm Prediction Center (SPC) is developing both a tornadic and severe nontornadic sample of supercell storms from 2014. This latest work is an extension of earlier research which led to the development of conditional probabilities of tornado damage rating from near-storm environment and radar-based storm-scale characteristics from a 5-year sample of tornadoes (4,770) reported in the contiguous United States (CONUS) during 2009–2013. The probabilities are derived from filtering tornado EF-scale segment data, large hail (i.e., ≥1 inch diameter), and severe wind gust (i.e., ≥50 kt) data by the maximum event type (e.g., tornado, hail, wind) per hour on a 40-km horizontal grid. Near-storm environment data, consisting primarily of supercell-related convective parameters from hourly objective mesoscale analysis calculated at the SPC, accompanied each grid-hour event. Filtered large-hail/wind events (~ 11000) associated with effective shear ≥20-kt and tornado events (800) were subsequently examined with level-II radar data. Convective mode was then assigned manually to each tornado large-hail/wind event if 0.5° velocity data from the nearest WSR-88D exhibited rotation. Peak 0.5° rotational velocity was recorded for each tornado event along the tornado path and within 10 minutes/miles for large-hail/wind reports. Preliminary results of tornado probabilities based on 2014 severe supercell 40-km grid-hour data are presented. Implications of these findings for diagnosing tornado potential in near realtime and possibly applying this research to aid National Weather Service (NWS) Impact-Based Warnings are discussed —tentatively scheduled for operational adoption NWS-wide by early 2016.