Identification of Ideal Smoothing Radii and Appropriate Seasonal Applications for Tornado Climatology Research

Tornado climatology research has shown that subjective, and sometimes arbitrary, decisions must be made when deciding what types of temporal and/or spatial bounds to apply to the data. Some studies include all events while others place importance on certain months, seasons, or periods of consistent frequency. Most researchers employ spatial smoothing techniques to overcome the relative lack of data in most locations, however, there is no clear “best” shape or size for smoothing methods.

The fundamental purpose of this research is to objectively determine the appropriate spatial extent of the bandwidth used to calculate tornado density values (i.e., smoothing the raw tornado data). Ultimately, the results show that the ideal bandwidth depends partially upon the temporal analysis period and the lengths of the tornadoes studied. Hence, there is not a “one-size-fits-all,” but the bandwidth can be quantitatively chosen for a given dataset. Results from this research, based upon tornado data for 1950–2010, yield ideal bandwidths ranging from 100 km, for the longest tornado paths and shortest periods, to 350 km for the shortest tornado paths and longest periods.

With the understanding that a smoothing radius depends partially upon the period of study, the next step is to objectively identify ideal periods of tornado analysis. Many previous attempts to define “tornado seasons” required arbitrary spatial boundaries and simple counts of tornado events within those boundaries. To avoid decisions about spatial or temporal boundaries, this project makes use of storm speed and tornado path length data, along with statistical cluster analysis, to establish tornado seasons that display significantly different patterns from one another. While spatial delimiters imply that certain locations must always or never be grouped with other locations, temporal delimiters allow locations to be included in the elevated risk area multiple times. Hence, while one area may be considered “different” from another region during the fall or winter, they may be quite similar during spring. This method yields five different seasons, with none longer than three months, with unique characteristics of storm speed and tornado path length.

The ideal smoothing radii are then applied via a kernel density analysis of each new tornado season. The resulting density maps illustrate clearly when and where tornadoes are most likely to affect locations across the contiguous United States. These density maps can be used to focus public awareness on times of the year when they are most at risk for tornado activity as some “high-risk” locations do not experience their temporal peak during the ubiquitous spring tornado season.