Illustrating Predictability for Nocturnal Tornado Events in the Southeastern United States

Tornadoes that occur at night pose particularly dangerous societal vulnerabilities, owing to the typically limited visibility at a time when the vast majority of people are sleeping or residing in weak-infrastructure buildings. Understanding these high-impact events is a crucial step for forecasters to provide life-saving warnings for public safety. This study builds upon previous research that assesses the ability for meteorological parameters to distinguish severe thunderstorm environments. In particular, this study uses the Statistical Severe Convective Risk Assessment Model (SSCRAM) to determine what parameters can be linked to tornado potential in the southeast United States during the months of November to May. This study shows that several parameters have statistically significantly different distributions between the Southeast and everywhere else in the contiguous United States, and between a coastal region subset of the Southeast and everywhere else in the contiguous United States outside of the Southeast. By specifying a constraint of at least 50 knots of effective bulk shear, the predictability for tornadoes in the southeast U.S. is generally better than everywhere else. Overall, the coastal region subset offers worse predictability, compared to an outside-of-Southeast domain, when no environmental constraints are added. These, and other distinguishing factors of meteorological parameters typifying the southeast U.S. November-May tornado environment, shed light on the predictability of this spatiotemporally focused regime.