Linking Spatial Patterns of Tornado Outbreak Intensity with Climatological Variability

Outbreaks of severe weather, including tornadoes, are challenging to link to climate-scale phenomena due in part to a lack of a formal outbreak definition, nonphysical trends in tornado data, and inconsistent methods of quantifying outbreak intensity. We present an updated definition of tornado outbreaks (TOs) in the United States based off spatially cohesive clusters of tornado reports using kernel density estimation; we then apply this approach to an updated index of tornado intensity. In developing this index, we use a support vector regression based detrending procedure to account for secular trends in tornado reporting, offering an updated approach to existing trend-removal methods from the tornado record. Using this intensity index to quantify climate-scale TO variability, we explore possible links between TO intensity and the leading modes of intraseasonal and interannual climate variability in North America: the El Niño Southern Oscillation, the North Atlantic Oscillation, the Arctic Oscillation, and the Pacific North American Oscillation. We will use point density analysis to explore potential spatial relationships between these teleconnections and TO frequency and intensity over the CONUS.