Poorly Warned Georgia Tornado

Last Saturday, an EF-2 intensity tornado injured five in far southeast Georgia. Unfortunately, the National Weather Service's (NWS) warning for this tornado was less than adequate. 

Radar Review

One of my goals in publishing these reviews is so that young meteorologists may benefit from my 53 years of radar experience in severe storms. So, we are going to review this storm in some detail. 

The above image is from 1:20pm when a circulation forms near Hoboken, Georgia. Rotation is shown in the right panel. From this point on, the storm should be closely watched, but I would not issue a tornado warning at this time. 

At 1:25pm, the rotation has tightened a bit and, with a "significant tornado parameter" of 1, a tornado warning should be considered, but I probably would have held off a bit longer to insure the rotation was well organized. 

At 1:28, the strong rotation has narrowed and I'm quite confident I would have issued a tornado warning at this time. The warning from the NWS had not yet been issued. 

At 1:31pm, there is no question in my mind that a tornado warning should have been issued. The rotation is continuing to increase in intensity and the reflectivity data (the type of radar seen on television and the left frame, above) is showing a more supercell-type look. 

At 1:33, the transition to the "supercell look" continues with little change in the rotation. The tornado will touch down in six minutes. 

1:36pm: There is a slight amount of "hollowing out" in the reflectivity data (upper left) which may mean raindrops are being flung out of the rotation, a sign a tornado is forming. The rotation is now focused on one point -- meaning a tornado is almost certain, especially since there is a high value of spectrum width in the same location (white line).

It was this data which prompted the NWS to issue a tornado warning which had a 1:37pm time stamp. 

By 1:41pm, this was obviously a strong tornado.

There was a hook echo in the reflectivity data, strong rotation, high spectrum width and, for the first time, lofted debris was detected by the Jacksonville radar. 

Images the from the local news media shows of the destroyed mobile homes and damaged conventional housing. 

The 1:37pm tornado warning gave just two minutes of lead time. That is not enough time, especially for the people that were in the mobile homes that were destroyed. In my opinion, it was possible to provide nine minutes of lead time.