Tornado Warning May Have Saved Lives

REBECCA ROBERTS, host:

People in Greensburg, Kansas, were warned about an approaching tornado on Friday night. Loud sirens sent many residents to shelter before the storm hit. Tornado watchers were able to issue that warning, in part, because of the work of Don Burgess, a meteorologist in Norman, Oklahoma. He works at the National Severe Storms Laboratory there.

Good Morning.

Mr. DON BURGESS (Meteorologist, National Severe Storms Laboratory): Good morning.

ROBERTS: So how much warning did residents in Greensburg, Kansas, have?

Mr. BURGESS: Well, they really had excellent lead-time. The specific warning that said a tornado was coming to Kiowa County, where Greensburg is located, was issued about 30 minutes before the tornado struck Greensburg. And the Dodge City National Weather Service Office then issued a tornado emergency for the town of Greensburg itself 10 to 15 minutes before the tornado struck.

ROBERTS: So we're to talking to you, in part, because of your work on the 1973 storm in Oklahoma. What can you tell us about that storm?

Mr. BURGESS: Well, we did research with a new tool, Doppler radar in the 1970s. It was also the early days of storm intercept, where we sent out scientists to observe storms. And our first big success was with the Union City storm of May 24th, 1973.

ROBERTS: And what were you able to learn from that storm?

Mr. BURGESS: Well, it was a very exciting day. We didn't have as good a forecasting as we have now, so we were first surprised to see a giant super-cell storm, and then even more surprised to start seeing signatures on our radar displays.

We could see the circulation forming up inside the cloud, the rotating precipitation particles, the rotating rain. Tornado genesis occurred after that, as much as 30 minutes after that. So we, for the first time, could give longer lead-times in tornado warnings.

ROBERTS: And in 1973 at that Union City storm, would 30 minutes have been unheard of as a warning time?

Mr. BURGESS: Yes, ma'am. At that time, the average lead-time was about two minutes. Also, the accuracy of the warnings was not very good, so people tended to not believe them.

ROBERTS: Kansans are prepared for tornados, more or less. They have warning systems. They have sirens in towns. They have shelters. Are you seeing more storms outside of Tornado Alley where there might not be that infrastructure in place?

Mr. BURGESS: Well, we've always seen some storms outside of Tornado Alley. Riegelwood, North Carolina, was struck back in November of 2006. Also struck at 6 AM in the morning, and that's a very difficult warning situation for the people of North Carolina, more difficult than the people in Kansas dealt with Friday night.

ROBERTS: So looking ahead for the non-meteorologists among us, what's the future of predicting storms? What would you have hoped that the residents of Greensburg could one day have that they didn't have this storm?

Mr. BURGESS: Well, I guess the first thing I would say is about 37 years ago now, when I got in this business, if I could have said that Greensburg got 10-15 minutes of warning saying your town is going to be hit, get underground, I would have been ecstatic because we did not have that. The longer lead-times are most necessary when we're dealing with sports venues or other gatherings of large people where you have to have time for them to disperse. So we really do need these lead-times up to an hour, where we can evacuate all facilities that might be in path of the tornado and get people to proper shelter.

ROBERTS: Thank you so much for joining us, Mr. Burgess.

Mr. BURGESS: You're welcome.

ROBERTS: Don Burgess has retired as chief of the Warning Research and Development Division of the National Severe Storms Laboratory in Norman, Oklahoma. He still works there on contract. Transcript provided by NPR, Copyright NPR.