Detection of severe Midwest thunderstorms using geosynchronous satellite data

Abstract

In the present exploration of the effectiveness of severe thunderstorm detection in the Midwestern region of the U.S. by means of approximately 5-min interval geosynchronous satellite data, thunderstorms are defined in IR data as points of relative minimum in brightness temperature T(B) having good time continuity and exhibiting a period of rapid growth. The four parameters of rate of T(B) decrease in the upper troposphere and stratosphere, isotherm expansion, and storm lifetime minimum T(B), are shown to be statistically related to the occurrence of severe weather on four case study days and are combined into a Thunderstorm Index which varies among values from 1 to 9. Storms rating higher than 6 have a much higher probability of severe weather reports, yielding a warning time lead of 15 min for hail and 30 min for the first tornado report.