Conditional Tornado Probabilities from RUC-2 Forecasts

Several previous studies have established statistical relationships between tornadic weather and environmental conditions by associating severe weather reports with rawinsonde observations. Here, we seek (1) to determine whether similar relationships hold when severe weather reports are associated with gridded short-term numerical forecasts, and (2) to develop and demonstrate a probabilistic model to forecast the likelihood a thunderstorm will be tornadic. Severe weather reports and lightning network data from 1 January 1999 through 29 June 1999 were used to classify the weather at a set of Rapid-Update Cycle (RUC-2) grid points into the mutually exclusive and collectively exhaustive events representing increasing severity of the weather. The events were no significant weather, thunderstorms, severe thunderstorms, or tornadoes. RUC-2 forecast convectively available potential energy (CAPE), helicity, and 0-6 km wind shear from these same dates were associated with this gridded classification of the weather. We generally found similar relationships between environmental parameters and storm severity as others have previously documented. Our Bayesian probabilistic model forecasts the likelihood of that a thunderstorm will be tornadic given a certain value of CAPE and helicity (or CAPE and wind shear). This model is shown to do a reasonable job of locating high-threat areas many hours in advance of the severe weather and may be of use as a forecast tool