The Climatology of Severe Thunderstorms: What We Can Know

Knowledge of severe thunderstorm occurrence is important for forecasters and researchers. High-quality climatological records can provide information that relates to temporal and spatial variability in forecast performance and atmospheric behavior. Unfortunately, most of what is contained in the various severe weather databases around the world is of little or no value for addressing those issues. As a community, we are caught between the horns of the dilemma of adequate sample size and adequate stable observational periods. I will illustrate some of the problems that exist both with the US and international databases, and will attempt to describe the most reliable aspects of the data.

At NSSL, an effort to define daily climatological probabilities based on the operational record has produced datasets that provide some information on the distribution of severe weather in the United States, although the data must be interpreted with many caveats. In general, we can have the most confidence in the distribution of days with significant (F2 or greater) tornadoes in the US. The areas that are the most prone to tornadoes, hail, and high winds can also be defined, but the magnitude of the threat is open to question. Distributions of tornadoes by damage scale show similarities around the world, with some regions (e.g., Florida, the Colorado Front Range, and the United Kingdom) being dominated by non-supercell processes, and others (e.g., the Great Plains, New England, France) dominated by supercell processes.

Limits on the detectability of forecast improvement and global climate change associated with severe thunderstorms will be addressed, keeping in mind Wittgenstein's imperative, "That of which we cannot speak, we must remain silent."