Using Statistical Learning Techniques to Investigate Spatial-Temporal Characteristics of Large Tornado Outbreaks in the United States
In this study we present a systematic analysis of tornadoes rated F2(EF2) or greater on the Fujita (Enhanced Fujita) scale which struck several US counties in one day. We investigated the trend and seasonality, clustering behaviors, geographical shifts and return period for high-strength tornado outbreaks during 1950-2020. The seasonality analysis revealed that 69% of large tornado outbreaks happened in March-April-May, and there was no record during August. Wavelet transformation of the time series revealed significant power at the decadal time scale during the 1970–1980 and a strong inter-annual cycle potentially linked to El-Niño Southern Oscillation. Spatial analysis of tornado outbreaks verified that the center of tornado outbreaks has been shifting to the Southeast US and Dixie Alley during the recent years compared to the earlier records. The variance of tornado clusters has significantly decreased in both latitude and longitude direction, which suggests a decrease in spatial dispersion. The findings related to spatial characteristics is of great importance given that the geographical shift could involve regions with a high density of mobile homes and forested terrain. Additionally, we used an exponential probability model for the return period of large tornado outbreaks and explored changes in the inter-arrival rates of tornado outbreaks during successive 30-year blocks and found that the arrival rate has changed from 124 days during 1950–1980 to 164 days during 1977–2007. The study provides valuable information for risk assessment of large tornado outbreaks at the US county level.