Engineering Tornado Intercepts - Dr. Joshua Wurman Discusses Mobile Doppler Radars

By James Turner
December 15, 2008 | Comments: 4

You may also download this file. Running time: 17:27

Subscribe to this podcast series via iTunes. Or, visit the O'Reilly Media area at iTunes to find other podcasts from O'Reilly.

James Turner: This is James Turner for O'Reilly Media. I'm speaking today with Dr. Joshua Wurman, the President of the Center for Severe Weather Research in Boulder, Colorado. Dr. Wurman has spent many years perfecting the science and technology of weather radar, technology that he uses on a regular basis now as he travels the country driving mobile Doppler radar units into places that most sane people are fleeing. The Center does research into everything from forest fires to hurricanes, but Dr. Wurman is probably best known from his participation in the Discovery Channel series Storm Chasers which chronicles the yearly attempts to study tornadoes in the center of the United States. Thank you for joining us.

Josh Wurman: Pleased to be here.

JT: So let's start with the inner-guts of the Doppler on Wheel or DOW itself; what exactly are you toting around on that truck?

JW: The Doppler on Wheels is a state-of-the-art research weather radar so we can collect data in tornadoes, hurricanes, anything that has particles in it that reflect. The difference that it's on wheels and we can drive it up close to things. A normal weather radar is usually so far away from an object that it collects a pretty blurry image and the only real way to get sharp image is just the same way as with your eyes; you've got to get up close to something. When we get up close we see tens of thousands better detail than a radar would that's 20 or 50 times further away.

JT: You've obviously got a radar unit up there but what have you got in the cab?

JW: In the cab we have three or four scientists who are driving, navigating and operating their equipment. Also we have a lot of communications gear and navigation gear because we're trying to conduct--really lead the choreography of the whole tornado intersect, so we're not just collecting our own data. We're also collecting the data off the internet and off satellite and then we're also directing other vehicles. So we need to know where they are, where they're going, where they're going to be in four minutes or ten minutes. We're constantly tracking where the tornado is both so we can intercept it but also keep our crews safe. Many of the vehicles have to get out of there and so we're tracking whether they have one minute, two minutes, three minutes to do their job and get out of the way.

JT: How many actual computers have you got onboard there?

JW: I think in the DOW we have about 10 or 15 different computers. Those range from very low-powered computers that aren't much more than laptops to very sophisticated computers that are hosting the signal processing for the radar, controlling our antennas and displaying and processing our radar data.

JT: And I assume that it's all being powered by a generator onboard?

JW: We have a 12 kilowatt generator and it powers the computers; it also powers air-conditioning because all those computers and people inside generate a lot of heat and we need to keep things cool. And also power the antenna; we have an eight-foot antenna that's spinning around and often very high winds. We can get winds over 100 MPH and we need to be pushing that antenna back and forth and still doing our mission even when we're in those hostile conditions. We scan that antenna as we're driving down the road; when we're driving down the road if we're going 70 or 75 into a 30 MPH headwind we have 100 MPH winds hitting the truck and hitting the antenna. So we need a lot of power in order to both operate the radar, operate the computers, air-conditioning and especially push the antenna back and forth in the wind.

JT: That kind of brings up an interesting question which is an antenna that big and a wind like that--there's a word for that and it's called a sail. How top heavy is that vehicle?

JW: Well the antenna is a big sail and we certainly can feel it as we're driving down the road when it's windy pushing the antenna and the vehicle back and forth. The antenna itself is pretty light and of course the weight of the sail doesn't matter so much. It's the aerodynamics of the sail. Our truck is quite heavy; our truck weighs about 25,000 pounds. It's very sturdy and we've designed it to have most of its weight low down, so we're very bottom heavy and can resist those kinds of motions. We operate in very hostile environments; we just came back from Hurricanes Gustav and Ike where we were right in the eye when they came onshore. We were on an overpass in Galveston and that radar is designed and has been very, very stable for us in a lot of hostile environments.

JT: Normally people shut their computers off when there's a thunderstorm nearby. You on the other hand are driving into the middle of them. Do you need to take special steps to protect the gear?

JW: Well I assume you're referring to lightning and lightning is certainly one of our concern, both for people because getting an electric shock or a burn from the lightning but also for our equipment. Our equipment is all inside these metal cages. The trucks are made of metal; we have heavy steel roll bar cages. We have steel mesh on some of the windows so the equipment is pretty resistant to nearby lightning strikes. We actually have had a lightning strike to the DOW itself in a thunderstorm and the equipment survived. We had to reboot the computers but we kept going. Our biggest concern for safety is when people are sometimes outside the vehicles, particularly if film crews are out there filming us from the outside and there's a lot of lightning around--even nearby strikes can knock you out or stop your heart. So we're concerned about that. We have crew members who know CPR; we carry a defib kit with us just in case somebody is hit. But we're always watching out for that.

JT: This year we saw the DOW go down for what looked to be the better part of a day--what ended up being a bad monitor cable. How difficult is it to keep all the pieces up and running in that?

JW: The DOW has a lot of parts and they're also built on what is a shoestring budget compared to most technology--its sophistication, you know. We don't have a Space Shuttle size budget or a military size budget, so we're doing a lot of our own engineering. I'm not an engineer; I'm a meteorologist but I am the chief engineer of our project. And it's a struggle to try to keep all those parts going and keep them going in a hostile environment where we're bouncing down the road and hitting water and wind and then suffering all those conditions. And sometimes it's sophisticated things that go wrong; something could be going wrong with some of our fancy signal processing, but in this case it was the most mundane thing. There was a pin in the monitor cable connector that was gone. It was one of the last things I would have thought of because usually monitor cables are a pretty secure thing. We're worried about other kinds of connectors and other specialized equipment in the radar that might be failing--not something as simple and mundane as a monitor cable. Obviously next time that will be higher on our list but next time it could be something else that's broken.

JT: Yeah; it's always the last thing you check.

JW: There's always something that's the last thing to check and the problem is with a system with this kind of sophistication is there's 100 possible failures. There are 100 possible failures and we just have to go down from most likely to least likely and usually it's the first or second thing we check, but every once in a while it's way down our list.

JT: How much data do you actually record during a typical day and how is it stored?

JW: Our typical day we record about 10 gigabytes of data with the radar and then other instruments actually collect a lot less; they're just collecting megabytes--weather instruments and things like that. We have a new radar system that can collect a lot more than that--probably 50 or 100 gigabytes in a day. Through the years we recorded that data in all sorts of different media. We used to use Exabyte types, then we moved onto CDs and then DVDs and now we're back just to external disks and then we take them out of the truck on big thumb drives. You can get 32-gigabyte thumb drives now and we just back up a day's data on that and get it into our lab.

JT: You place probes in what you hope is going to be the path of the tornado. How do you insure that you're actually going to have any data left after it gets hit?

JW: Well there are thousand and some tornadoes in the US every year. Only five or ten-percent of those have the worst winds where you're really getting winds of 200-300 MPH; so first of all 90-percent of those intercept(s) are going to hit tornadoes where the probes can survive. Even with those five or ten-percent it's only a part of that tornado that has the worst winds, so if they're getting hit by the fringe of the tornado or typically what's the north-side, the weaker side of the tornado they might not get extremely high winds that are capable of really shredding one of those pods. The last thing is that even if we were so lucky that we really were--or so skilled, depending on how you look at it, to get one of those pods in the strong part of a violent tornado even though the instruments on that pod are destroyed, the data logger, the computer that's recording all that data is almost indestructible. It's in a small heavy steel case that's stronger than a black box in an aircraft that's supposed to survive an airplane crash. So even if that pod got toppled over or lofted or thrown into a ditch, it's got this waterproof heavily armored steel case where the data is being recorded. So we think we would get all the data up until the point that the weather instruments were destroyed. But as I said, you know we should be so lucky that we would get that kind of impact. It would be a very successful mission if that happened; I would love to have an instrument destroyed in a tornado. It would mean I did really well.

JT: You spend your days sitting a couple of feet from a radar unit. Excuse my ignorance, but why don't you get fried?

JW: Well a radar--a weather radar typically doesn't really put out that much energy and the energy is pretty distributed. So for example, my radar sends out 250 kilowatt pulses, but for a very short duration, so the average power that's going out of my radar is about 200 watts. That power is distributed over an eight-foot antenna, so even if I stood right in front of the radar beam I couldn't even get warmed up because that's 250 watts distributed over a huge area, so it's not like a light bulb. A light bulb has 100 watts and it's very, very hot; you don't want to touch it because it's coming out of a very small area. The last thing is nobody stands right in front of the beam for hours; that would be a pretty silly thing to do. So there are several reasons why a typical weather radar can't really hurt you. It's possible that there are military radars which have higher intensities and stuff which really could cause some harm if you stood right in front of the beams but we measure the radiation that comes out of our system and it's just not that intense. It's not much more than really what's coming out of the sun on a strong sunny day.

JT: Another question in a similar line; when you've got multiple Doppler units operating in the same area for instance you and the NEXRAD or multiple DOWs how do you avoid getting each other echoes back?

JW: Well we operate at a very, very different frequency than the NEXRAD. We're operating at nine-something gigahertz and they're operating at two point something gigahertz, so they're far apart. But there are sometimes issues with radars that are operating in the same band interfering with each other and if there are other research radars out there we need to be taking some care to avoid the frequencies that they are and kind of tune to a slightly different frequency. Just like television stations or radio stations do; they operate within the band but slightly different frequencies.

JT: You also study hurricanes; what's the benefit of a DOW in a hurricane since hurricanes tend to be around for fairly long periods of time? They've got aircraft flying through them; they've got ground observations. They've got the regional radars going off. What are you bringing to the table there?

JW: Well there's still big questions about hurricanes and what the winds are like at the lowest levels. When we took the first DOW into a hurricane, Hurricane Fran back in 1996, it really was what we call a fishing expedition. We didn't know exactly what we would pull up but we discovered a whole phenomenon which has given us a pretty major new understanding about the energetic(s) of the boundary layer, the lowest levels in hurricanes. We discovered the winds aren't just blowing in a straight line, but they're kind of going in a corkscrew. There are these rolls happening in the lowest levels of the hurricane. And those probably have a very important role in how hurricanes intensify and weaken offshore. And one of the major--probably the major outstanding questions in hurricane research and hurricane forecasting is to be able to predict when and by how much hurricanes are going to intensify or weaken. We pretty much know the track for the hurricane; these forecasts are pretty good but knowing whether it's going to hit shore as a cat-one or a cat-four is hard to say 72 hours in advance. And so there are still a lot of things to learn that we can only learn by getting up close, by getting into the storm with something that can collect all those fine scaled measurements.

JT: What have you learned over the past few years about tornadoes and severe thunderstorms and what's the current Holy Grail of data that you'd like to acquire?

JW: Well we've learned a lot about the structure of the tornadoes. We've learned or probably put boundaries or fenceposts on what our ignorance is which is very important and helps us define how to answer more focused questions. One of the major outstanding questions we have about tornadoes is when exactly they're going to form in the super cell, how strong they're going to be, and the difference is between severe rotating super cells that make tornadoes and those that don't. Most severe rotating thunderstorms don't make tornadoes and there are subtleties that we don't really understand about the structure of those storms or about the environment that they're going into which are affecting whether they make no tornadoes, weak tornadoes or the very few that make strong tornadoes. And we think it's some combination of the structure of the storm and the temperatures of the air in and outside of the storm that are making those differences. And we hope that by surrounding these super cells with multiple radars and in situ instruments and cars with instruments that we can learn enough about the interaction between the temperatures in the environment, the precipitation field, the rain and hail in a storm and the wind fields in those storms to know which particular structures--which particular environments are likely to make strong tornadoes as compared to the majority of environments which result in a no tornado at all.

JT: Well speaking of places that normally don't get tornadoes, this past summer we had a reported F-1 come through the seacoast of New Hampshire which is something I can't remember happening in my lifetime. Is there any evidence that tornadoes are becoming more severe or appearing in more diverse locations?

JW: No; there's alot of speculation about whether climate change will affect the distribution or intensity of tornadoes, but there's very little hard science behind any kind of prediction. The problem is the conditions that make tornadoes are fairly subtle. If you just get a warmer environment it doesn't necessarily result in more tornadoes. If you just get a cooler environment it doesn't necessarily result in less tornadoes. Tornadoes typically happen in the Midwest; they happen in the springtime--not during the hottest times. Tornadoes don't happen in the hottest countries. If you think about it, Brazil is pretty hot and tropical; Saudi Arabia is pretty hot and dry--neither of those have tornadoes--or they don't have many tornadoes. Similarly, tornadoes aren't happening in the peak of the summer in Oklahoma when it's 105-degrees every day; they're happening in the springtime or if a subtle combination of features are coming together to make these tornadoes. So every once in a while you get tornadoes in New England. One of the famous tornadoes in US history was the Worcester tornado in Massachusetts back in the 1950s; so they can happen there. It's just they happen less frequently.

JT: Okay; I'm going to ask one question about the show itself. Having a film crew around must have something of a Heisenberg effect. How much does the presence of the Discovery crew factor into the day-to-day operations of the team?

JW: Well fortunately the Heisenberg effect of the crew is pretty minor. It has almost no impact on the weather of course and we have an arrangement and ground rules with the Discovery Channel and their production crew that basically mean we do our science unencumbered by what they are doing and they just film us. It seems to be a zoo out there; there are more of them than there are of us. And we take 30 hotel rooms a night and there's a fleet of a dozen cars out there because there are always production crews following us. But aside from that kind of effect, we really have a very good relationship with the Discovery Channel and their production company that lets us do what we want to do; let's us do our science. I would say the main Heisenberg of having the Discovery Channel is out there is that by supporting these efforts and by funding these efforts more science is getting done. It's not the traditional source of scientific funding which is usually for us the National Science Foundation but in between years when we really can't get continuous funding from agencies like that for these types of experiments, the Discovery Channel has filled that gap and let us learn more about tornadoes than we otherwise would have known.

JT: All right; well we've been talking to Dr. Joshua Wurman, President of the Center for Severe Weather Research. Thank you for taking the time to talk to us and hopefully we'll see you again next tornado season.

JW: Okay; thank you very much.


You might also be interested in:

4 Comments

Nice to see some technology behind the DOW upclose, these kinds of things aren't always showcased in the TV broadcasts.

Not sure how big your crew is now Josh, but after that monitor cable incident it does sound like you could use a dedicated computer tech. When you strip off all the advanced equipment your still sporting computers and someone should be there dedicated to taking care of them for you.

Some of those probes at their core seem pretty small, i've wondered why you guys don't buy a tiny RC chopper and fly/launch the probes up at the tornadoes.

Hopefully next season will be a good one for you guys, look forward to seeing you on TV, its an awesome series

I have to say I'm a bit surprised at some of the description. Radar will pass through lots of things that air won't, so it would seem relatively easy to set up something transparent to the weather Radar set that deflected the air around the antenna. That would improve the stability of the vehicle (rather important for the people inside), the stability of the antenna itself (which would improve data collection) and reduce turbulence (the main reason equipment in such a system fails). Presumably there;s a good reason they don't use such an approach, they're extremely bright people with plenty of resources, but the reasoning escapes me and wasn't covered in the interview.


Likewise, on-board data processing doesn't sound so smart. You can't do ADC with any accuracy if your reference voltage isn't absolutely perfect, and voltages in a vehicle are never clean. Vehicles riding through tornadoes less so. Again, there must be a very obvious (to the scientists) reason why they don't just record the raw signal and replay it under better conditions. I would have thought analogue collection and clean processing would give greater accuracy than in-situ digital collection, but they think otherwise and this is their field, not mine. I would have liked the interview to have gone deeper into the reasons for their choices.


I understand podcasts can't get into all the gritty detail, simply for time reasons. This is one of the problems with any interview of scientists or engineers. The subject simply requires more depth than space permits, leaving readers/listeners intrigued but not necessarily very informed. This is as good as any interview I've seen, and kudos to O'Reilly for including it. There needs to be some way of carrying supplemental information, though, or a greater enthusiasm for follow-up interviews. I don't know which would be better. Maybe a bit of both.

jw how can I sign up to chase storms

How long time will you "... collect data in tornadoes..." Josh? Tornado have unlimited number of datas, similar like trees in a wood have unlimited number of foliants. But if you can't see roots behind foliants, you can't understand nothig about trees! You see lot of data and can't see basics. As much data as worse understanding. Can you understand it Josh? High tech will not help you. Russian mind can.

News Topics

Recommended for You

Got a Question?