´óÏó´«Ã½

NARRATOR (DILLY BARLOW): This is a story of science against nature at its most furious. It is the battle to understand what triggers the most terrifying and destructive tornadoes, the Supertwisters.

Dr JOSH WURMAN (Centre for Severe Weather Research, Colorado): A two hundred and fifty or a three hundred mile an hour wind can only be compared to those that might be experienced on the fringes of a nuclear explosion.

Prof HOWIE BLUESTEIN (University of Oklahoma): It may be strengthening, and there may be another tornado possibly. Got to get back in, there's a very, very strong in flow in to the storm right now.

NARRATOR: At stake is something that could not be more important, people's lives. Hundreds are killed and injured each year by these Supertwisters. For any one living where these tornadoes strike, any extra warning time could quite literally save lives. Tonight, Horizon tells of the struggle to find a way of giving those few precious extra minutes warning, that might not save properties or possessions, but could offer people just enough time to reach shelter and safety.

NARRATOR: In 2002, a small town on the east coast of America received an unwelcome visitor.

CONTRIBUTOR: It's coming this way guys. I'm leaving.

CONTRIBUTOR: I remember seeing this huge funnel cloud. And it was holding things in the air that were just whirling so fast. And as they came to the top you could recognise things, like a tree, a couch, a desk.

CONTRIBUTOR: And we're looking at each other thinking is that a tornado. Where's it going.

NARRATOR: It was a giant tornado. And it swept straight through the heart of La Plata, Maryland.

CONTRIBUTOR: And I remember this tangled mess of metal was coming right at me, I mean directly at my car. And briefly I thought oh I'll probably get decapitated. And my car lifted two feet up in the air, and actually moved to under the CVS sign.

CONTRIBUTOR: Still I, I say it sounded like a jet, turbine engine, I can't even duplicate the sound. The house started rumbling, shaking.

NARRATOR: The monster tore through the town.

BARBARA WATSON (National Weather Service): This tornado was moving close to sixty miles an hour, that's a mile a minute from, from here over to the buildings that are standing there. The tornado was probably through here in fifteen seconds, or less. It's amazing, you go from almost no wind up to two hundred mile an hour wind and back down.

NARRATOR: Yet as the tornado was ripping through La Plata, just outside of town, Susan Erikson and her husband Eric were oblivious to the unfolding disaster. They were visiting the construction site of their new dream house.

SUSAN ERIKSON: The only thing we had heard on the news that day was thunderstorms, and we had thought no big deal. We had just come up from the basement, I remember silence, at that time no birds, no anything, I just heard silence. We were getting ready to leave and then everything went blank.

NARRATOR: Susan came too in a pile of rubble, and heard the groans of her husband Eric.

SUSAN ERIKSON: Both my arms were broke, they weren't functioning correctly. So what I did was try to manoeuvre my legs so that I could stand up and drag my arms up with me. I remember walking out of the house.

NARRATOR: Some neighbours found Susan before she passed out again. When she was recovering in hospital, they gave her news of her husband.

SUSAN ERIKSON: It was between the two surgeries before I went in for my orthopaedic surgery on my arms that they told me he had died. The rescue team they said I shouldn't have lived, I really should have died with Eric because I was directly behind him. There is no rhyme or reason why I was left behind, I just truly don't know if I was blown back where he was caught under the rubble or what exactly happened.

NARRATOR: Eric Erikson was one of five people killed that day, by one of nature's most violent forces. Because what had struck the town of La Plata was no normal tornado. It was a Supertwister. Some violent storms, like hurricanes, are so huge they cover thousands of square miles. They can even be seen from space. There can be literally days of warning before they strike. But tornadoes are altogether different. Compared to hurricanes they cover a much smaller area, and they form at incredible speed, it is this that makes them so deadly. Tornadoes are measured on a five point F scale. The weaker ones are classed as F zeros, or F ones. Yet even these can reach speeds of over a hundred miles an hour. And it's not just the force of the wind that makes tornadoes so feared, but what the wind can carry with it. Hailstones the size of golf balls can wreck a car, and kill a person. At a university in Lubbock, Texas, scientists have recreated the power of a tornado using compressed air, looking at the impact of missiles travelling at a hundred miles per hour on bricks, plate glass and sheet metal. They hurl the kind of debris that would be picked up by a tornado and assess the damage they would cause. Few would survive a missile travelling through the air at a hundred miles an hour, and that can be just a normal tornado, but in some cases winds are much, much greater. Rated F four and F five on a five point scale, these are the deadliest of tornadoes.

GARY ENGLAND (Meteorologist, Oklahoma): They do happen and they happen nearly every year, and the force of an F five comes down your street, if you're not in a safe room or basement or cellar you, you're probably going to be history.

NARRATOR: These rare beasts are known as Supertwisters. They can unleash enormous energy and can generate ground level winds about two hundred miles per hour. Josh Wurman from the Centre for Severe Weather Research in Colorado is a world expert in Supertwisters.

Dr JOSH WURMAN: The winds of strong tornadoes are among the strongest forces that we ever see in nature. The over pressures that are experienced by buildings during a two hundred and fifty or three hundred mile an hour wind can only be compared to those that might be experienced on the fringes of a nuclear explosion.

NARRATOR: In May 1999, Wurman measured the fastest Supertwister ever recorded. It first set down in a corn field, about forty miles south west of Oklahoma City, a city of more than half a million people. Unusually for a tornado, TV forecaster Gary England had a full hour to warn his audience.

GARY ENGLAND: Do not try to ride this storm out in your home unless you are trapped. Get in the centre part of your house, a tub in your bathroom, cover it with pillows and blankets, lots of pillows, lots of blankets, get in the bathtub, put the kids in the bathtub, get in on top of the kids.

NARRATOR: From Channel 9's command centre, England and his viewers could see exactly what was coming.

GARY ENGLAND: Right now it may turn a little bit north of Norman, see if it maintains itself.

TIM MARSHALL (Meteorologist): They served us with continuous coverage, they had helicopters up in the air filming the tornado so people could watch the TV, see exactly where the tornado was. They had radar with fantastic capability showing the path that this thing was going to take. So people knew what was coming.

GARY ENGLAND: We have tornado on the ground, right there, right inside. Tornado on the ground...

GARY ENGLAND: May 3rd, you know we had nearly seventy tornadoes in our viewing area here in Oklahoma. And you know a lot of people killed, what eight thousand structures destroyed, people pay attention now.

NARRATOR: Out on the road Josh Wurman got close enough to measure the wind speed. It was an astonishing three hundred and one miles per hour. It made the Oklahoma Supertwister the most powerful ever recorded.

TIM MARSHALL: A three hundred mile an hour wind is not three times as strong as a hundred miles an hour. It is nine times as strong.

Dr JOSH WURMAN: They can cause such complete destruction, and then it's apparently almost instantaneous, one moment you're out there on an afternoon, and twenty minutes later your house can be gone.

NARRATOR: A wind travelling at three hundred miles per hour would be capable of flattening practically anything in its path. This power makes the Supertwister especially dangerous. The only way to guarantee survival is not to be there when they strike, just a few minutes extra warning time could mean the difference between reaching safety or death. And finding those few extra minutes could be all the more vital because there's something making the situation even more critical. Cities in Tornado Alley are becoming bigger targets.

TIM MARSHALL: I think we have reached a critical point here, where as we are now expanding our cities bigger and bigger, so the targets for tornado hits are bigger and bigger. And the fear that I have is that now that the city is going to continue to get bigger, the houses still stay the same in terms of construction, now we're going to start having an upswing in the number of fatalities.

GARY ENGLAND: So let's just picture this, let's say it's two in the afternoon, largest tornado comes in to a city. Moving towards a school that has five hundred children in it. What's going to happen when that warning goes out? There's going to be hundreds of parents trying to rush to that school to get those kids out. And it will take the school out, it will take the parents out. And mark my words it will happen.

NARRATOR: It's fears like this that put huge pressure on scientists to come up with a solution, an effective way of predicting Supertwisters. If an answer could be found then hundreds of lives would be saved. The first line of defence against severe weather is the National Storm Prediction Centre in Norman, Oklahoma. All Supertwisters are produced by thunderstorms. But so far it has proved impossible to predict which storms will go critical and which ones won't.

STEVE WEISS (Storm Prediction Centre, Oklahoma): There are several limitations to our knowledge at this point. And the reason for that is that while we have made some progress in understanding physical processes on the scale of thunderstorms, we don't understand nearly enough from a scientific perspective. We all wish that we could pin point the exact location where severe weather will strike, and tell people what time it's going to strike. But the science hasn't reached that point yet.

NARRATOR: The storm centre monitors the atmosphere across the whole of North America, and then uses computer programs to try to predict where the most dangerous storms will develop. And if the general conditions look strong for tornadoes, the storm centre will issue a so-called tornado watch. But a tornado watch can cover tens of thousands of square miles, areas so wide that these warnings are often viewed as meaningless. Only after radar sees a tornado is actually forming, does a precise tornado warning go out.

GARY ENGLAND: This is a major tornado, wind speed we don't know.

NARRATOR: But by then it can often be too late.

GARY ENGLAND: Do not try to ride this storm out in your home unless you are trapped.

BARBARA WATSON: When we issue the warning, the warning is supposed to mean that that tornado has been detected, it is there. We want to be able to take it to that next step where we can say we've got this really bad storm, conditions are right for a tornado to develop. Where we're at right now, when you've detected it it's occurring.

NARRATOR: The sad fact is that official tornado warnings often come too late. On average only twelve minutes before a Supertwister strikes. Not enough time to evacuate a school or hospital. Finding out what turns a thunderstorm in to a Supertwister had become a vital scientific challenge. This has become the mission for a very select band of scientists, known as the storm chasers. Howie Bluestein has been chasing tornadoes for over twenty-five years in the twister prone reason of America's central plains, known as Tornado Alley. When he started out Howie's operation was certainly lo-tech. There were no laptop computers, no global positioning system, nor mobile phones. But over time the storm chasers' arsenal has expanded.

Prof HOWIE BLUESTEIN: Twenty-five years ago we went out by ourselves. Now when we go out there is an armada of cars out there. There are a number of different mobile Doppler radars. And of course the more vehicles you have out there the more vehicles that need gas, and more people that have to run to the bathroom at bad times.

NARRATOR: Today the storm chasers main weapon is a Doppler radar dish mounted on the back of a truck. Doppler radar uses sound waves to sense the movement of air and moisture in remarkable detail, even picking up patterns of dust clouds and insects. This equipment is the most accurate way of mapping the complex swirling wind patterns that make up a tornado. Today storm chasing teams often include at least two of these Doppler trucks so they can record the storm from different angles. Howie was the first to put a Doppler radar dish on the back of a truck, and catch data off tornadoes in action. The storm chasers hope that all this data collecting equipment will one day help crack the problem of predicting Supertwisters.

Prof HOWIE BLUESTEIN: What we want to do is capture the formation of tornadoes in many, many storms. We want to see precisely how the wind field is changing, we want to see what's happening to the temperature field and the humidity field for a lot of storms.

NARRATOR: A big problem is this equipment doesn't like to travel, and when you're storm chasing you have to be prepared for anything.

STORM CHASER: There's a horse in front of us.

STORM CHASER: Oh look at that.

NARRATOR: They face treacherous road conditions, dust storms, and hail the size of golf balls.

Dr JOSH WURMAN: Radars at some kind of level aren't really designed to be put on trucks and bounced around in severe storms, and so we have a lot of problems with the complicated electronics, where things just shake and break apart, and we have shorts, things like that happening all the time.

Dr JOSH WURMAN: Not much I can do about that.

Dr JOSH WURMAN: So we've broken our windshields now for the last two days, our last one only lasted seven hours, that's very frustrating.

NARRATOR: The simple fact is that tornadoes are so unpredictable tracking them is still hit or miss. Storm chasers like Howie Bluestein consider themselves lucky if they intercept one or two good tornadoes each year. With such a haphazard way of collecting information it has always seemed likely that the Supertwister will hang on to its secrets. Those few vital extra minutes of warning time would remain elusive and lives would continue to be lost. But one man thought he might have a solution. Kelvin Droegemeier's aim was to use a new computer model to provide early warnings, not of Supertwisters themselves, but of the conditions that spawn them.

Prof KELVIN DROEGEMEIER (University of Oklahoma): My dream is to be able to detect tornadoes down at very fine scales, to anticipate tornadoes, a half an hour before they occur. We're talking about a storm forming to produce a tornado over a country or a city, it's a whole different ball game.

NARRATOR: Droegemeier's starting point was the knowledge that Supertwisters are all formed under the bellies of thunderstorms, in particular a hugely violent type known as a super cell. A super cell is a vast rotating column of air. These huge bodies can be twenty miles across and sixty thousand feet high. More than double the height of Mount Everest. Every Spring, warm moist air surges up from the Gulf of Mexico, pushing in to cool dry air from the north. Strong winds coming in from different directions at different speeds causes the air to start rotating. And as the energy becomes more intense, a super cell is born. But only in some of these super cells will an angry tail emerge to reach down and touch the ground. The main problem is understanding this final step, what is the precise mechanism by which a super cell gives birth to a Supertwister?

Dr JOSH WURMAN: Somehow super cell, thunderstorms, are able to bring intense rotation and intensify that rotation near the surface. And we don't understand that process very well. We know that before a tornado forms there are areas or rotation in the thunderstorm. What we don't understand is how that rotation is brought down to the ground and intensify in a very short period of time in order to make a tornado.

NARRATOR: In the search for a trigger one of the prime suspects has been a burst of air descending from the storm called a downdraft. A blast of air that descends and picks up speed as it falls towards the ground.

Dr LOUIS WICKER (National Severe Storms Laboratory, Oklahoma): We think the downdraft is really important for producing the tornado because it helps sort of squeeze the air from near the ground up in to the updraft. The downdraft kind of comes around and squeezes the air in the front and the back together, it kind of squirts the air up off the ground.

NARRATOR: In the minutes before a twister forms, storm chasers have often seen a downdraft blasting a hole in the clouds, toward the rear of the storm.

Dr JOSH WURMAN: That's a very impressive storm, but the structure looks quite good, we're seeing a very strong RFD coming down. But at this part there's still no tornado.

NARRATOR: But just as often, even when a super cell forms a tornado fails to materialise. For all their long years of study, storm chasers are forced to admit that they still haven't worked out the exact chain of events that turns a thunderstorm in to a tornado.

Dr JOSH WURMAN: We know that super cell thunderstorms make tornadoes, but we also know that most of them don't. Only about twenty, or twenty five percent of super cell thunderstorms produce tornadoes. And only perhaps one to five percent of those produce what we call significant tornadoes, the large ones, the ones that do ninety, ninety-five percent of the damage and fatalities.

STORM CHASER: I feel like we're in a Twister movie.

NARRATOR: The problem lies in the complex interactions that make the weather. Weather systems are some of the most intricate patterns known to science.

Prof HOWIE BLUESTEIN: Whether a thunderstorm forms in, in one county or the next county, could depend upon differences in the wind that you can barely detect with instruments. Or changes in temperature or humidity that are barely detectable.

NARRATOR: And Howie knows just how far we are from understanding the true nature of tornadoes.

Prof HOWIE BLUESTEIN: We still haven't solved the problem yet, we still haven't figured out exactly why tornadoes form.

NARRATOR: And this is why Kelvin Droegemeier's approach offers a potential way forwards. His idea is not reliant on knowing the precise details of why a tornado forms, he believes that the answer lies in finding a way of more accurately predicting thunderstorms. His aim is to create a revolution in forecasting, Droegemeier has spent fifteen years building a powerful computer program that will breakdown the behaviour of the weather.

Prof KELVIN DROEGEMEIER: I think that the advances in science and technology are taking us in a natural direction toward a new mode of doing forecasting. Because right now the models don't predict thunderstorms at all, they predict general areas of precipitation, but they can't predict say an individual thunderstorm firing out in north western Oklahoma, the kind that, that we see sweep across the plains every year.

NARRATOR: In his system Droegemeier planned to use a mass of Doppler information, but in a completely new way. Previously meteorologists have used Doppler images to track storms after they form. But Droegemeier was hoping to use this data to help predict thunderstorms hours before they appeared. He has developed a powerful computer model that combined this Doppler information with temperature and humidity data from satellites and weather stations, which he hoped would accurately predict thunderstorms for the first time. Last year in the middle of the new tornado season he was given the perfect opportunity to test it out. On the morning of May 8th the weather began to turn. By mid afternoon the National Storm Prediction Centre had blanketed large portions of the southern plains with tornado watches covering tens of thousands of square miles. So Droegemeier set his computer model to work. He predicted a severe storm in a particular location in Kansas. And five hours later tornadoes were born. Droegemeier's computer model had predicted a thunderstorm in that very place. it was a level of precision normally unheard of.

Prof KELVIN DROEGEMEIER: Very excited that part of the model did really well, up in south eastern Kansas, it gave a signature of very strong storms in south east Kansas, many hours before the storms actually occurred.

NARRATOR: Yet on the same day, part of the model did not perform so well. In the next state another Supertwister had formed and struck the town of Maura, Oklahoma. This time the tornado made a beeline for Mama Lu's Diner.

CONTRIBUTOR: Oh we were sitting out here having dinner and my daughter called and said that a tornado was coming our way.

CONTRIBUTOR: The manager told us we either had to leave or stay.

CONTRIBUTOR: Things happened so fast and people moved so fast, and the man in the wheelchair, we couldn't get this door shut.

CONTRIBUTOR: When the tree came through the hallway we just fell kind of backward.

CONTRIBUTOR: When we walked out the walls were gone, everything was gone, the ceiling was in. There were people laying everywhere, on the highway and in the ditches. But we were all alive, it was amazing, amazing.

NARRATOR: Just a few minutes later the storm tore the side off the local General Motors plant, and levelled the Union Hall. It was a massive event. But for the hour that the tornado struck, just after five pm, Droegemeier's model had got it completely wrong. It had predicted clear skies over the region.

Prof KELVIN DROEGEMEIER: Central Oklahoma though where we had a very isolated event that we to this day don't know what triggered that thunderstorm, the model didn't do well at all.

NARRATOR: Given these difficulties, some worried that Droegemeier's approach will ever be useful in predicting tornadoes.

Prof HOWIE BLUESTEIN: You might be able to predict that a storm may or may not form, you may predict that some storms may be more likely to produce strong tornadoes than others. But we may never be able to predict that any given storm, any given location, will go on to produce a tornado.

NARRATOR: Droegemeier though is not put off. He believes it's all about collecting more Doppler information. So he's building an experimental network of small dishes, twenty miles apart on mobile phone towers and buildings across Tornado Alley.

Prof KELVIN DROEGEMEIER: We want to put a few of these radars out there, to make sure that if a tornado was beginning to form, we know absolutely for certain that it's going to happen. And we can tell you that thirty minutes ahead of time whether it's a weak tornado, a strong tornado, whether it's in December or, or March, or May.

NARRATOR: Droegemeier's technique could one day lead to great success. In the meantime, others may have made a significant breakthrough. Another Spring, another chance for the storm chasers to collect more vital data. May 15th, 2003, was perfect weather for tornadoes. Storm chaser Josh Wurman and Howie Bluestein led separate teams. What they discover could one day prove vital, but their experience has also showed just how unpredictable tracking tornadoes could be.

Prof HOWIE BLUESTEIN: May 15th was incredibly complicated. We, and everyone else in the world recognised that over the Texas Panhandle the conditions were ripe for tornadic storms, the conditions looked very, very good.

NARRATOR: For a storm chaser very, very good conditions meant that wind from different directions was mixing to create an environment ripe for super cells. A zone of low pressure drew warm moist air from the south in to the northern part of Texas, known as the Panhandle. Just above it, a current of air called the jet stream was blowing even faster. The two air currents, crossing at different speeds and altitudes, caused the air near the ground to begin to spin. If there was a thunderstorm in the area it would take on the spin, and perhaps go on to spawn a tornado. As the chase started the team leaders had to chose which part of the Texas Panhandle looked most promising. A lot rode on the decision.

Dr JOSH WURMAN: Well we have a lot of anxiety because it's a very high stakes game, it's very expensive and time consuming to go out after these storms and they're fairly rare. So we really need to make these calls efficiently if we're going to capture these rare events. And we only have a few chances every year to do that.

NARRATOR: Josh decided to head to the northern part of the Panhandle towards Dalhart, Texas. But Howie was unsure which route to take. To help him make up his mind he decided to get the most up to date information available, so he found an internet connection on route.

Prof HOWIE BLUESTEIN: Oh man, oh my god will you look at this. Look at the back winds and they're strong, oh wow, oh it's going to be Amarillo...

NARRATOR: Howie, unlike Josh, thought the southern Panhandle held more promise. He also feared that up north, where Josh was heading, the road system was too sparse, meaning they would not be able to get close enough to get a good reading. So Howie chose to stay near Amarillo, Texas, and wait for the storms to come to him. He waited, and he waited, but nothing happened. Meanwhile in the north, Josh Wurman suddenly got lucky. When the wind touched down he was in the perfect position.

Dr JOSH WURMAN: We intercepted six tornadoes on that day. And one thing that was fascinating about that storm was that there were at least two occasions where there were two vigorous tornadoes on the ground at the same time.

NARRATOR: It was a goldmine of information, and the first time that twin tornadoes had ever been captured with mobile radar.

Dr JOSH WURMAN: Having two tornadoes in close proximity like that, maintaining their vigour, is very unusual. We'd never seen that before in all our years of radar intercept.

NARRATOR: Josh and his team had had a fantastic day, they had captured new evidence on multiple tornadoes, including rare twin twisters. In the meantime, about a hundred miles to the south, the weather was depressingly fine. And Howie had captured nothing.

Prof HOWIE BLUESTEIN: It's a storm chasers nightmare to know that he made a good forecast in general but there's a tornadic storm going on that you could have gotten, that's just a hundred miles away. You've driven all this, this great distance, made what you thought was a really good forecast, and, and you've missed a tornado, you feel terrible.

NARRATOR: Things were looking bleak, but after hours of waiting around, watching the clouds on radar, Howie finally got an encouraging phone call.

Prof HOWIE BLUESTEIN: Hello, yes Don, we are sitting in Hedley right now watching it on our radar. Fantastic, that's what we've been watching, ok thanks a lot. Ok, um I'm going to come and have a look at it, we have some good news. Oh my god, wow.

NARRATOR: A thunderstorm was forming right behind him.

Prof HOWIE BLUESTEIN: Wow that looks, that looks good, that really nice.

NARRATOR: What Howie saw was the vertical edge of a thunderstorm, pushing up through the atmosphere, and beginning to rotate, a super cell.

Prof HOWIE BLUESTEIN: All of a sudden one of the storms suddenly went up and became a super cell, as if someone had turned the switch on.

Prof HOWIE BLUESTEIN: It looks like a good super cell, we're going to have to move here, we're having trouble with the computer.

NARRATOR: Finally, a familiar pattern began to appear on the radar screen. The winds seemed to be producing the distinctive hook shape often associated with a strong swirling downdraft and a tornado.

Prof HOWIE BLUESTEIN: Er we have the hook, it's just to our west, north west, probably right now less than ten miles, and it's er, it may be strengthening and there may be another tornado possibly. We're collecting data right now.

NARRATOR: Hunting a tornado after dark was extremely dangerous.

Prof HOWIE BLUESTEIN: I hate to not be able to see it. If we didn't have the radar we wouldn't be here. Better get back in there's very, very strong in flow in to the storm right now.

Prof HOWIE BLUESTEIN: The storm had been moving at thirty-five miles an hour, coming right towards you, and you were near the storm but you didn't know precisely where the storm was with respect to where you were, you could have been in great danger.

NARRATOR: But the Doppler radar managed to pierce the night, and the team captured the kind of valuable information that could one day lead to a vital breakthrough in understanding the Supertwister. Because Doppler data like that captured by Howie and Josh is being used by Lou Wicker and his colleagues, at the University of Illinois. They have spent a decade developing a way of working out precisely how a tornado actually forms.

Dr LOUIS WICKER: It's very likely that in order to solve the tornado problem we're going to need a lot more data than we have right now. Because a lot of things are going on at very small scales that we don't actually sample. Now we're moving in to an area, we're trying to understand the dynamics of storms. Dynamics means how the individual blobs of air are all interacting with each other to produce the strong winds or even the tornado.

NARRATOR: Unlike Droegemeier's computer model, that set out to predict thunderstorms before they happened, Lou Wicker wanted to show how a thunderstorm actually became a tornado. Attempting to visualise a storm using real tornado data had never been done before.

Dr LOUIS WICKER: The closest approximation we have are computer simulations, and we try to build a digital computational model of the tornado. And see if we can change variables and have storms that produce tornadoes and ones that don't.

NARRATOR: Wind speeds, the atmospheric pressure and humidity, all observed before a real tornado had taken place, were input in to the computer model. The virtual storm grew by itself, and Wicker watched to see whether the data he had input was enough to create a tornado. The model showed the clouds spin and churn high in the sky. Just as they did in nature, these clouds descended closer to the ground, still swirling rapidly. The question was, would it generate a tornado. He then noticed the rotation suddenly narrow and intensify. Wicker began to see a pattern. Warm air, shown in orange, rose in to the storm until it slammed in to the jet stream above. This appeared to make the column of air spin rapidly. And then, a tornado was born. But it begged one key question, what had triggered the tornado. The first clue was something familiar. The simulation showed a clear hook of wind, and with it a strong downdraft.

Dr LOUIS WICKER: Very warm moist air mass in this region here, and that's, that's very interesting because that means that while there is a sort of a downdraft in there it's not a cold rainy downdraft, it's a sort of warm juicy downdraft which will really help feed the storm.

NARRATOR: But the experiment revealed more. In addition to the downdraft Wicker saw something else. The model showed that just as the storm intensified a series of whirlwinds formed at ground level. These tiny corkscrews of wind merged together in to a much larger current, and seemed to form the tornado itself.

Dr LOUIS WICKER: Very chaotic and it's very complicated, it's not just a simple little downdraft there and just spins itself up, there are lots of things going on at the same time. I think it captures a lot of the complexities that we see in the real world.

NARRATOR: Wicker's model seemed to offer an insight in to the intricacy of what sparks a twister. One of the triggers for tornadoes could be those tiny whirlwinds. And if a way could be found to identify these tiny whirlwinds before Supertwisters are formed, it could be vital for prediction.

Dr JOSH WURMAN: If we could increase warning times by five or ten minutes that's really significant. If we could push back the forecast of a tornado to fifteen minutes before the tornado forms, people have fifteen minutes to get to their shelters. So small increases of just fifteen minutes, twenty minutes, can have a great impact on the value of that forecast.

NARRATOR: The whirlwinds that showed up in the computer were so subtle that in real life they'd be invisible to the naked eye. So far this was all theoretical, until recently this had not been seen for real. But Howie Bluestein thinks he's seen evidence of this phenomenon.

Prof HOWIE BLUESTEIN: We've been able to actually catch tornadoes in the act of forming with very, very high resolution. And what you see is a lot of small vortices. You don't see one vortex becoming a tornado, you see a number of small vortices.

NARRATOR: So Wicker's model seemed to work. Mini whirlwinds had shown up on both Howie's and Josh's radar before the tornado had formed. If someone could find a way of identifying these subtle patterns early enough, it could be a key to finding extra time in tornado prediction.

Prof HOWIE BLUESTEIN: I think that the solution maybe in sight. We, we will I think in the next, within the next ten years, with the mobile Doppler radars, actually see why tornadoes form.

NARRATOR: It is early days, but the science behind what causes a Supertwister really does seem to be moving forwards. And everyone is painfully aware that this progress can not come too quickly.

Prof HOWIE BLUESTEIN: We have been with tornadoes that have gone through towns and have killed people, and it makes you, makes you feel sick, it's a horrible, horrible feeling. But at the end of the day we feel that perhaps on the basis of what we've learned, over a number of years, that we'll be able to eventually give people better warnings and hopefully save lives in the future.

NARRATOR: Every year twisters reek havoc across the world. Back in April, one claimed the lives of eight people, only three minutes of warning were given. It is for this reason, that at the start of each year, the tornado chasers' quest continues with a renewed sense of urgency.

Back to top

Back to the Horizon homepage