Analysis of Unwarned Tornado Events from 2014-2015 across the National Weather Service Central Region (40)
Fred Glass, NOAA/NWS, St. Charles, MO
Rod Donavon, NOAA/NWS
Jacob Beitlich, NOAA/NWS
Aaron Johnson, NOAA/NWS
Kevin Deitsch, NOAA/NWS
The National Weather Service Central Region Tornado Warning Improvement Project (TWIP) Team examined unwarned tornado events from 2014-2015 to determine if there were any common factors that contributed to the lack of advanced warning, and ascertain potential training needs for warning meteorologists. A total of 484 unwarned tornadoes were identified using Storm Data from the NWS Performance Management Database. After extensive quality control between 340-447 tornadoes remained for various forms of more exhaustive analysis. The majority of the unwarned tornadoes were weak (97%), short-lived, and narrow with a median path length of 1 nm, a median path width of 50 yards, and median lifespan of approximately 3 minutes. Events were filtered by convective mode with supercells the most prevalent across the entire NWS Central Region Domain at 40% and linear modes slightly less at 37%. Examining the geographical distribution of events by convective mode, linear modes dominated the unwarned tornado events within the Mississippi Valley Region while supercells dominated the Great Plains. WSR-88D radar data was analyzed utilizing Gibson Ridge GR2Analyst software to diagnose mesocyclone and mesovortex characteristics using storm relative velocity and Normalized Rotation (NROT) data. Detection of tornadic debris signatures were also reviewed. Hourly mesoanalysis data developed by the Storm Prediction Center was used to evaluate the near storm environment. External factors that may impact the warning decision making process including convective watch status and prior tornadoes on event day were considered. Finally, a fairly significant number of events in the original pre-quality control dataset had spatial or temporal errors in Storm Data that resulted in false positive identification of unwarned tornadoes. The Storm Data errors had a negative contribution to the Government Performance and Results Act Goals.