The Impact of Damage Indicator Density on Southern Plains Tornado Intensity

An attempt is made to quantify the impact the damage survey tornado intensity underrate bias has on the large difference between tornado intensity climatology derived via damage surveys and mobile Doppler radar data in the Southern Plains of the United States. The distribution of tornado intensities in the Southern Plains, like the rest of the country, skews heavily toward EF0 tornadoes with progressively fewer strong (EF2/3) and violent (EF4/5) tornadoes. However, a mobile Doppler radar climatology of 50+ tornadoes in the same region found that the preferred tornado intensity was EF2, and that the frequency of strong and violent (weak) tornadoes was much greater (less) than that found using damage surveys. It is well documented that there is an underrate bias when tornado intensity is estimated via damage surveys owing to the lack of reliable damage indicators that can be used to provide a wind speed range estimate and EF scale rating for many cases. In this study, an attempt is made to quantify that bias using a simple measure: the number of damage indicators per unit length of the tornado path length in the National Weather Service Damage Assessment Toolkit (DAT). The distribution of tornado intensities is then mapped as a function of this damage indicator density for all 2016 Southern Plains DAT cases. Preliminary findings and several caveats to this methodology are discussed.