Scientists Chase Tornadoes to Solve Mysteries Scientists still don't understand why some supercell thunderstorms form twisters and others don't. Ira Flatow and guests discuss the new VORTEX2 project, in which scientists hope to solve twister mysteries by using mobile weather stations to chase tornadoes around the plains.
NPR logo

Scientists Chase Tornadoes to Solve Mysteries

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript
Scientists Chase Tornadoes to Solve Mysteries

Scientists Chase Tornadoes to Solve Mysteries

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript


This is TALK OF THE NATION SCIENCE FRIDAY, I'm Ira Flatow. We're broadcasting live from the Reynolds Performing Arts Center at the University of Oklahoma in Norman, and on campus here is the National Weather Center. I visited this yesterday.

It's in an enormous complex, combining high technology and brain power, where they track, they predict, they research storms, tornadoes, weather of all sorts. And when I visited this center yesterday, I saw a huge fleet of mobile weather vehicles all ready, just getting ready to take to the road.

There's everything from big semis with 10-foot-tall radars on the truck bed to minivans with all sorts of barometers and anemometers and all kinds of wacky-looking stuff mounted on their roofs, complete with - oh, and some of these vans are complete with golf-ball sized dents.

If you're in Oklahoma, you know what I'm talking about, or in Minnesota, whatever, from - and even broken windshields that come from nice particles of hail, big golf-ball-sized hail.

And as I say, in less than a month, this fleet of weather vehicles is going to be chasing tornadoes all over the Great Plains. That's part of the Vortex2 Project. We'll be talking about that this hour.

We'll also be talking about - what's the point of storm-chasing? It's not just for fun or entertainment on cable news, is it? There are a lot of unsolved mysteries about tornadoes, like why do some storms spawn twisters, and how come there are no tornadoes in other storms? Why do some tornadoes last longer than others? What's actually going on in the clouds? That's something that we really don't know a lot about. And how can we get better at predicting tornadoes?

So we'll be talking about a lot of these things today. Our number, 1-800-989-8255. You can tweet us @S-C - @scifri, @-S-C-I-F-R-I, and also if you are here in the audience, there are a couple of microphones. Please feel free to step up to the microphones and ask your questions. Let me go to our Web site and remind you to go there, and you can go to and get more information about what we're talking about this hour - also our new video that'll be up there as we speak.

Don Burgess is a research scientist at the University of Oklahoma here in Norman. He's former director of the Warning Research and Development Division at NOAA's National Severe Storms Laboratory. He's also a Steering Committee member on the Vortex2 Project. Welcome to SCIENCE FRIDAY.

Mr. DON BURGESS (National Severe Storms Laboratory): Thank you, glad to be here.

FLATOW: You're welcome.

(Soundbite of applause)

FLATOW: Greg Carbin is the warning coordination meteorologist at the NOAA Storm Prediction Center here in Norman. Welcome to SCIENCE FRIDAY, Mr. Carbin.

Mr. GREG CARBIN (NOAA Storm Prediction Center): Thank you, Ira.

(Soundbite of applause)

FLATOW: Don, let's start off talking about - I mentioned what we know and what we don't know about tornadoes. Give us a little thumbnail sketch of what we know, what we dont know.

Mr. BURGESS: Well, there's a number of things we do know about the storms. Most of those have come since World War II. We've used radar, other instrumentation, and we've started to go to the field and study these storms.

So we know that the rotating updraft of a special kind of storm we call a super-cell is the one that produces the most dangerous tornadoes.

FLATOW: When you say rotating updraft, what's going on? The wind's going up and down, or what's happening?

Mr. BURGESS: Well, the - there is a vertical change to the horizontal wind.

FLATOW: So the wind's usually blowing on the ground sideways, right?

Mr. BURGESS: The wind's blowing sideways, and it gets stronger as you go aloft, and in an updraft, that gets tilted. That vertical change of wind gets tilted, and so there's a propeller on the ground going like this, and that gets tilted, and that becomes a horizontal rotation at the mid-levels of the storm. And radars can see that well, and we've seen that for some time, and we use that for warnings, and we now, because of that in part, issue better warnings.

FLATOW: So what don't we know then about the...

Mr. BURGESS: Well, you gave a nice list of things that we don't know.

(Soundbite of laughter)

FLATOW: And why don't we know them?

Mr. BURGESS: Well, they're hard to observe is I guess the best answer to that question. We used to think if we got rotation right at the ground, we'd probably get a tornado. In Vortex1 in 1994, 1995, we learned that there is some strong rotation right at the ground, not just in the mid-levels, yet we don't get a tornado. So there's something special that goes on, and we have to translate the rotation that maybe is aloft down to the ground.

So downdrafts can be important in that process. They're hard to observe. We don't want to fly through the storms. We can make measurements at the ground, but even those can be somewhat dangerous. So we struggle to see everything that we need to see. We keep learning little pieces. We're peeling back layers of an onion, but it's a very complicated onion.

FLATOW: Well, we'll get back to this in a little bit. Let me talk to Greg Carbin about the, what happens day to day in the Storm Prediction Center. You're there - what goes on in that?

Mr. CARBIN: Well, what we're trying to do is foresee these, the plains and other parts of the country, before they form, and given that we, as Don just described, have such a difficult time seeing some aspects of the storm even after they form, you can imagine how difficult it is to see the possibility that these storms will begin in the first place.

And so our role, 24/7, 365 days a year, is kind of like the outlook. You know, we're watching the sky. We're watching the country for the potential for these severe storms to form because they're devastating once they form, and we're looking one, two, three days out.

FLATOW: And so you make predictions based on what you see coming on all these computers that you have going on there?

Mr. CARBIN: Yes, and we kind of assess the pattern of the atmosphere and the ingredients that are required for thunderstorm formation, and then once we do see the potential for storms in a certain area, we kind of focus in on that area and look at it a little closer, try to assess the ingredients more accurately.

FLATOW: Which are? What ingredients?

Mr. CARBIN: Moisture.

FLATOW: Coming from?

Mr. CARBIN: The humidity in the air, moisture from water sources like the Gulf of Mexico. Instability, which is essentially buoyancy in the atmosphere. When you see a cumulous cloud, there's instability associated with that. And then lift, something to lift the atmosphere up.

And those occur on a variety of scales, but when you put them together in the right way, with the right wind shear in the atmosphere, you can get a pretty amazing storm.

FLATOW: What is there about the geography of this area, the Tornado Alley, that makes these ingredients happen? What's happening here?

Mr. CARBIN: Well, it is geography, and it's topography, the fact that the Rocky Mountains are just to our west, and the Gulf of Mexico is off to the southeast. We have westerly winds in the atmosphere that create an area of low pressure, kind of almost semi-permanent to the - downwind of the Rocky Mountains in the plains. That low pressure draws air northward from the Gulf of Mexico, and we have the right setup for storms if you're given these other ingredients.

So we have moisture, mountains, plains. Everything's in the right place.

FLATOW: And it's just the job of making the forecast, the tiny little job.

Mr. CARBIN: Yeah, sure.

(Soundbite of laughter)

FLATOW: Well, I was struck when I was visiting, we were chatting yesterday, and you were showing me all of the technology that you have there, and I visited different people, and what struck me is that despite all the electronics that you have, it's somebody's call.

Mr. CARBIN: It is.

FLATOW: It's like the umpire's going to make that call. You know, there's no computer going to make that forecast.

Mr. CARBIN: There's still a little art in the science of meteorology, and there's still gut instinct on occasion, pattern recognition. Experience place a role. I've been here before, I remember this, and this is how things evolved when I saw this pattern. There's no doubt that that plays an important role in the forecast process.

But we have a tremendous amount of information coming in too, from satellites to radar data to the computer models that we can integrate into that experience as well.

FLATOW: Don, tell us about Vortex2. I mentioned this fleet of trucks and things and sensors. What are they going to do?

Mr. BURGESS: Well, to get this data that's hard to get, as what we established, we need to go to the storm, not wait for the storm to come to us. So we take all of our instruments and drive them underneath the storm, and our radar is at the edge of the storm, and actually deploy pods ahead of the tornado to try to get the most detailed measurement.

So we take out this whole armada, this array of instruments that you described, and we try to study one storm in detail and understand in very fine resolution what's going on with that storm.

FLATOW: Is this before the cloud, the tornado cloud forms, or you look for it and then chase after it?

Mr. BURGESS: No, we need to be there before it starts. If we want to understand what we call tornado genesis, the process of formation, we need to be there earlier. Now, this is after the cloud has formed. If we do what we're supposed to be doing, we're in the right region, and it might be one cloud or another cloud, and once we see one developing strongly, then we all converge on that one storm and we collect our data.

FLATOW: Of course, you can't leave the equipment in the tornado as it's occurring, can you? I mean, you don't want to lose that - it will get lost, right?

Mr. BURGESS: We have some pods that we put in front of the tornado, and we might lose one, but actually it's harder to get one right in front of the tornado than you might imagine if you keep a degree of safety when you deploy them in advance.

So most of our instruments are guided by people, and they get out of the way, and these pods, if we lose them, so be it, but last year, we didn't lose any.

FLATOW: And you also have drones that are lying around, right?

Mr. BURGESS: We have UAVs, unmanned aerial vehicles. We can't fly those in our entire domain where we collect data for some reasons, but there is an area, a rural area where we can fly those, and they're very useful. You wouldn't want to fly a manned aircraft in and around the tornado, but you can fly a UAV there, and we can get very good information.

FLATOW: With all this - being in the age of such satellite technology, why can't we use, why can't we use satellites to collect...

Mr. BURGESS: Well, it's a matter of scale, really. We're looking with a satellite at something the size of a small city. You can't really drill down much smaller than that. A tornado can be on a much smaller scale, and so we need something - what we call in situ, right in place to observe some of these things.

FLATOW: And Don, so then you want to get right up into that cloud because we can't collect the data and we haven't seen what's in that cloud before.

Mr. BURGESS: Right.

FLATOW: And we don't know really what's going on in there.

Mr. BURGESS: We don't know the fine details, particularly at low levels, and so that's why we drive all these vehicles, we fly the UAVs, we take the radars and bring them up very close, and we have different wavelength radars that have different missions on scanning the whole storm or scanning the area of rotation or maybe just trying to scan the tornado.

FLATOW: Now, I imagine when you get these trucks out on the road, and they're -you know, it's something like we've seen in the movies, the movie "Twister," whatever that is, you must have people waiting for you to go out and chasing with you, eh? Do they get in your way?

Mr. CARBIN: Or go to their basements.

(Soundbite of laughter)

Mr. BURGESS: In this day and age there are a number of people who like to chase storms.

FLATOW: Right.

Mr. BURGESS: That's become something that's popular among at least a certain group. And so when we go out and try and collect our data, we do encounter people that are out there, and we don't caravan all together as we approach the storm, since we have different missions.

We can't spread out, but occasionally people do follow us, and there's already people who are there, and we try to work with them, but we also ask, as much as we can, for them to help us by getting out of our way and at least giving us a place to park by the side of the storm.

(Soundbite of laughter)

FLATOW: It's that crowded?

Mr. BURGESS: It sometimes gets that crowded.

FLATOW: Wow. We're going to take a break and come back and talk lots more with Don Burgess and Greg Carbin. Our number, 1-800-989-8255. We have a couple of microphones out here, set up to take your questions, talking about tornadoes right here in Oklahoma. So stay with us. We'll be right back.

(Soundbite of music)

FLATOW: You're listening to SCIENCE FRIDAY from NPR. I'm Ira Flatow. We're in Norman, Oklahoma, talking about tornadoes this part of the hour, with Don Burgess, who is a research scientist at the University of Oklahoma and Greg Carbin, who is at the NOAA Storm Prediction Center here in Norman. Our number, 1-800-989-8255. A question from the audience here?

Unidentified Man #1 (Audience Member): Yeah, hi. My question is about Oklahoma and tornadoes in general. I've lived here almost all my life, and of course, people from other states go, well, tornadoes here. Why the hell don't you just move away from them?

But more recently, I've heard a lot more reportage of tornadoes all over the U.S., in Ohio and all sorts of places. And I'm wondering, is that a recent phenomenon, or is that just an aspect of reportage?

Mr. CARBIN: It's an aspect of reportage for the most part. The attention that some of these programs and research, but mostly Hollywood, have given to tornadoes and severe weather, have brought people out, and the Weather Service, the National Weather Service, including trainspotters throughout the country to report severe weather back to their National Weather Service office, and we're just seeing an increase in the number of people involved in that endeavor. So that's probably what you're seeing.

FLATOW: Thanks.

Unidentified Man #1: Thanks.

FLATOW: Let's go to the phones, to John(ph) in Index, Washington, is it, John?

JOHN (Caller): Yeah, that's it.

FLATOW: Go ahead.

JOHN: Well, I think maybe your screener should've worked a little harder here, but how about firing bottle rockets, you know, with sensors of some sort? You know, you've got the drones, and that sounds like you've got it covered.

FLATOW: All right, let's see if we can get an answer. Thanks for calling because both Don and Greg are shaking their heads on this. Don, you want to take it?

Mr. BURGESS: We've tried rockets and shooting them from light aircraft near the storm to try to hit the tornado.

FLATOW: And attempt to do what, break it up?

(Soundbite of laughter)

Mr. BURGESS: No, no, not to break it up. The - although some farmers out there thought we were trying to do that.

(Soundbite of laughter)

Mr. BURGESS: Thought we were practicing weather modification. All we're really trying to do is get more information with sensors you can put in the rocket. But it's a very turbulent environment. It's very difficult for the plane to fly, and actually, it was very hard to try and aim those rockets.

These were the small rockets. The much larger rockets you could probably get through there, but the big rocket's going to land someplace, and you might have a negative effect from that.

(Soundbite of laughter)

FLATOW: Yeah, I remember watching cloud-seeding experiments 30 years ago in Florida, and they had to stop that because the public was wondering what was going on.

If you had - let me give you the blank-check question. If you had a blank check to create an instrument or some technology that you don't have, and you think you would really need, what would it be? What would you like to have to study the tornadoes or the clouds better? Greg, do you have an answer while Don's thinking about it?

Mr. CARBIN: Well, I think if we had more computing capability, I mean, we're using some of the fastest computers in the world to simulate the atmosphere on a day-to-day basis, but we're still short when it comes to trying a variety of different approaches in terms of the computer models that we have, and it's really limited by computing capacity and computers - these computers - are not cheap. So I need a blank check for that.

Mr. BURGESS: I'm an observationalist. So I'm going to say more observation and particularly above the ground. We get a lot of information right at the ground, including in Vortex2, but we struggle to get the temperature, dew point, wind field, precisely above the ground.

The UAVs help. Weather balloons, which we use in Vortex2, help. But we just can't get all of the data everywhere nearly as well as we do at the ground.

CONAN: How good are you now at predicting a tornado happening. Has it gotten longer and longer?

Mr. CARBIN: It's improved dramatically, actually, in just a relatively short term. In terms of the large-scale conditions favorable for thunderstorm development, there have been amazing advances in the last decade and a half or so, especially with the use of simulations and computer models. Our ability to understand those conditions when storms will form have improved quite a bit.

Now, the weather changes from day to day, and some patterns are more predictable than others, and when we're in the winter season and early spring, we're usually in a pattern where, if we're going to see a large outbreak of tornadoes, we're going to see that a couple, few days in advance.

When we get into summer, the pattern gets a little less predictable, and we may not have a lot of lead time. So it varies during the year.

FLATOW: So what would be an ideal prediction you would like to make, an hour or two hours, six hours?

Mr. CARBIN: Well, our tornado watches are designed to be out two hours prior to the occurrence of the tornado event. We don't always do that.

FLATOW: Don't always do that. Okay, let's go to this gentleman right here. Yes.

Unidentified Man #2 (Audience Member): Here in central Norman, we don't have a large of occurrence of tornadoes tracking directly across the town, but we see several going to the south through Purcell, Noble and particularly more in the Edmund area, but would the curvature of the Canadian River cause a tornado to track differently?

Mr. CARBIN: There's no evidence of that.

Mr. BURGESS: There's a lot of energy that's associated with that storm, and to disrupt it is going to take an equal component of energy and apply it somehow. Even more hilly, mountainous areas are susceptible to tornadoes. People would like to believe or think they're protected by a river or a hill or something, but...

Mr. CARBIN: Indian legend.

Mr. BURGESS: An Indian legend.

(Soundbite of laughter)

Mr. BURGESS: I wouldn't count on that if there's a storm coming.

FLATOW: All right, thank you. Here's a question from Second Life, where folks are gathering - interesting question - and it comes from LaCursorFactor(ph), a scientist. How do El Nino and La Nina affect Oklahoma's weather or just tornadoes in general? Let me expand that.

Mr. CARBIN: There are connections. The connections are somewhat tenuous, but it does appear as if - well, the first thing we're talking about here with El Nino and La Nina are rather large-scale features of the globe, Pacific Ocean causing impacts on the jet stream. And so we're looking on a different scale at those features, and when we get down to the scale of a tornado, that's a much different environment.

So it's difficult to make a strong connection from one to the other. There have been studies to suggest that the La Nina, which is the cool phase of the oscillation, has in the winter months and early spring played a role in perhaps more active severe weather.

El Nino, and we're in one right now, and we've been in El Nino through the winter months, may be a suppressor in some areas, but at the same time, there have been El Nino events during the cool season, in the winter, that have ended up with devastating tornado outbreaks along Florida.

So Florida's had susceptibility to tornadoes during El Nino years, but this year they did not. So there may be connections there, but they're not solid, and they're not the only thing going on.

FLATOW: Let's go down here, yes.

Unidentified Man #3 (Audience Member): Hi. I was wondering if there was a scale that measured the strength of tornadoes, like...?

Mr. BURGESS: Yes, there is. It's named after a famous researcher, Dr. Ted Fujita from the University of Chicago. It's called the Fujita Scale, and in recent years, we have enhanced that scale by adding new information and things that have been learned since the 1970s, when the scale was enacted.

So we now have in place what's called the Enhanced Fujita Scale to measure tornado intensity, wind speed and the related damage that occurs.

FLATOW: So remember that, think of it in Spanish as fajita.

(Soundbite of laughter)

FLATOW: Now you'll never forget Fujita because I've made that association, tacos or something like that. That's how I remember these things, 1-800-989-8255. Because of global warming, will that affect the tornado season at all or anything?

Mr. CARBIN: It's possible. There are still uncertainties with respect to the tornado record, and our first speaker, question here was interesting because he asked: Is there an increase in tornado numbers? Well, there's an increase in the reporting of tornadoes, but our record, as you go back in time, is not a very good record.

The tornadoes in the early part of the century were the strong ones, the powerful ones, and they occurred where people lived. You need a human being to experience these events.

So we have to understand the underlying problems of using that record and drawing conclusions from it, and we just don't have the time, you know, to do that.

FLATOW: Don, speaking of the record, you've been digging into the history of the famous tri-state tornado.


FLATOW: I'm sure that means something different here than it would be back in New York, tri-state. So tell us what that is.

Mr. BURGESS: It's a famous tornado occurrence, arguably the worst tornado we've ever experienced in the United States, occurred March the 18th of 1925. The three states that were affected were Missouri, Illinois and Indiana.

And myself and some colleagues here, in our older years, are kind of taking on this research project as a labor of love because we're interested, but it's very hard to get the data from time long ago, and we've used a number of sources.

I don't have the time to go into it, but we've really done a study with more modern scientific thought, and we're just at the place now where we're producing drafts and going to publish our results, and I think what we've learned is yes, that was a very bad tornado, and probably, at least with everything we can see, was pretty much continuous across this tri-state area, over 225 miles of length.

FLATOW: Two hundred and twenty-five mile tornado.


FLATOW: Wow. I guess you have to find people who were around and saw it back then, right?

Mr. BURGESS: Well, in part that's what we do. We can get data otherwise, but we've done an extensive search for survivors, although they're in their 80s and 90s. This research has been going on for about 10 years now, and we have interviewed about 50 survivors of the tornado who've told us their story, and we've worked with a lot of descendents of survivors to get information.

FLATOW: So they can point out to a spot where the tornado went, right, that you wouldn't be able to find out otherwise.

Mr. BURGESS: Right.

FLATOW: There used to be a general store in that spot.

Mr. BURGESS: We tracked a piece of property by piece of property for the entire length of the track.

FLATOW: Wow. That's terrific. Well, we've run out of time about the tornadoes. I want to thank you, Don - and Greg can hang around a little bit. Don, thank you very much and good luck. When are you going to be publishing that because we'd like to know about it, a little bit later?

Mr. BURGESS: The (unintelligible)


Mr. BURGESS: It's - as a labor of love, it's one of those things we work on and then we have to go back to our real job occasionally. And I'm getting ready to go on VORTEX2 and we hope to get a lot of data and publish it and get better applications and better warnings from that. But I think we should see some of the Tri-State articles appearing within the next year.

FLATOW: Well, thank you, Don Burgess for taking time to be with us.

Mr. BURGESS: Thank you.

FLATOW: Don Burgess, research scientist at the University of Oklahoma...

(Soundbite of applause)

FLATOW: ...former director of the Warning Research and Development Division at NOAA's National Severe Storms Laboratory. He's also in the steering committee for the VORTEX2 project.

Copyright © 2010 NPR. All rights reserved. Visit our website terms of use and permissions pages at for further information.

NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.