Associating U.S. monthly tornado activity with environmental parameters

The analysis of observations in the vicinity of severe thunderstorms and tornadoes has led to empirical associations between the likelihood of tornado occurrence and local synoptic environmental variables such as convective available potential energy and vertical shear. These relations provide guidance in the production of probabilistic convective outlooks. We have recently used relations between monthly averaged environmental parameters from the North American Regional Reanalysis (NARR) and monthly tornado activity from the NCDC Storm Event database to construct an empirical “tornado index.” The index is a function of atmospheric parameters and is designed to represent the expected monthly tornado activity conditional on the monthly averaged local environment. The identification of associations between tornado activity and environmental parameters on seasonal or monthly time-scales is highly desirable since such associations would help to explain observed variability and could provide a path toward extended-range prediction of tornado activity. We find that the tornado index is able to reproduce aspects of climatological and interannual variability. Analysis of the index computed with forecast parameters gives some evidence of the predictability of monthly tornado activity.