Shear parameter thresholds for forecasting tornadic thunderstorms in northern and central California

A study of 39 nontornadic and 30 tornadic thunderstorms (composing 25 tornado "events," as defined in the text) that occurred in northern and central California during the period 1990-94 shows that stratification of the stronger tornadic events (associated with F1 or greater tornadoes) on the basis of 0-1- and 0-6-km positive and bulk shear magnitudes is justified statistically. Shear values for the weaker F0 events could not be distinguished statistically from the "background" values calculated for the nontornadic (null) thunderstorm events observed during the period. Shear magnitudes calculated for the F1/F2 events suggest that these tornadoes had developed in an environment supportive of supercell convection. Hindcasting the tornado events based upon shear thresholds produced a high probability of detection (POD) and low false alarm ratio (FAR), particularly for the stronger (F1/F2) events. Although the current sample size is limited and the conclusions drawn from it should be considered preliminary, it appears that California forecasters may be able to use shear profiles to distinguish days on which there is a higher threat of storms producing moderate and significant tornadoes. Buoyancy, as indicated by surface-based convective available potential energy (SBCAPE), was weak for each of the categories, and there were no statistically significant differences between SBCAPE values for each of the categories. Thus, as is true elsewhere, buoyancy magnitude alone appears to be of no value in forecasting whether California thunderstorms will be tornadic.