A Storm Is Born

The wind crosses the plains like a steady gust from a blast furnace. The red terrain, flat and featureless, can't divert it. The few trees that inhabit the landscape lean permanently to the north. In the distance, giant Oklahoma and United States flags are pegged, rippling as they extend northward in a perfect line.

fluffy clouds in skyJames Marquis

Clouds rise minutes before a storm initiates above meteorologist Paul Markowski and his research team.

The winds naturally blow in moisture from the Gulf of Mexico,” says meteorologist Paul Markowski. “That's what makes this a great area for storms.” But the sky, bright blue and stretching endlessly behind a Philips 66 gas station and a crumbling diner called the Outpost, doesn't promise much of anything.

It's a Wednesday in May. 83 degrees. Markowski, an assistant professor of meteorology at Penn State, is cooling his heels with a couple dozen students in a dusty parking lot, leaning against a Ford Contour topped by an aluminum rack 7 feet tall, fitted with plastic tubes and a propeller. A device that looks like a safety shower hangs just above the car's windshield. These strange-looking contraptions measure perfectly normal phenomena—wind speed, dew point, pressure, temperature, and geographical coordinates—but they seem to raise a fear of the supernatural in some of the local people. “Are you hunting for aliens?” one man asks. Another: “Is that a wind-powered car?” Other passersby are a little savvier. They've seen storm chasers with similar rigs. They ask: “Are we going to get tornadoes today?”

“Geek mobile” is what this car is called. Officially, it's Probe 8, one of a dozen or so vehicles in this afternoon's deployment. The parking lot is full of similarly tricked-out cars and vans with lots of twenty-some-things lounging inside of them, leaning against them, tossing balls back and forth over their hoods. Markowski, only 27 and boyish-looking in his shorts, T-shirt, and sandals, hardly stands out as a faculty member as he moves from car to car, talking and joking with his students. Until, that is, he starts explaining atmospheric phenomena.

He tells them now that a dryline—an invisible boundary line between dry air from the southwest and moister air from the east—is setting up nearby, outside of Guymon in the Oklahoma panhandle. The formation of this dryline, and its slow eastward movement, might be enough to spark a storm, he explains. He and the students are waiting for instructions from field coordinators Erik Rasmussen and Conrad Ziegler, research associates with the National Severe Storms Laboratory. Today, their job is to orchestrate the movements of nine probes, four radar trucks, three balloon-launching trucks, two research aircraft, and a camera truck. They do it all—with an earful of input from the radar coordinators, the aircraft crew, and several other scientists—from captain's chairs in front of huge computers inside a white van called FC, for Field Coordinator.

The “armada,” as the scientists like to call it, is part of a major research project called International H2OProject (IHOP), the largest field experiment in atmospheric science ever conducted in North America. IHOP includes people and instruments from research institutions such as the National Severe Storms Laboratory, the National Center for Atmospheric Research (NCAR), Penn State, the University of Oklahoma, and several others. The lead scientists are Dave Parsons and Tammy Weckwerth of NCAR. Penn State meteorologists Markowski, Ken Davis, and Yvette Richardson, who is one of the radar coordinators, are conducting several experiments with their students throughout the six weeks that IHOP is running. Stationed in Norman, Oklahoma, and in Liberal, Kansas, the researchers make almost daily forays onto the plains of Oklahoma, Texas, and Kansas.

grass and sky with long white cloudJames Marquis

Long flat clouds, part of a distant storm system, stretch across a clear blue sky.

On the other side of the parking lot, FC's antenna is raised. Rasmussen is downloading the latest weather data off the Internet, compiling images that will show him the exact location of the dryline. The plan is to set up the radars at the corners of an imaginary box 12 miles on each side, through which the dryline will pass. The probes will canvass the area inside the box, collecting data across the boundary as it moves.

Rasmussen speaks excitedly over the radio, his voice high and tight. He urges everyone to stay close to the vehicles. “We'll be leaving in the next ten minutes, as soon as we get the latest information from Norman.”

Twenty minutes pass. Rasmussen gives another warning not to leave the vehicles. A few stragglers run from the gas station with chips, sodas, and packages of Little Debbie Snack Cakes.

Finally, Rasmussen calls out coordinates, and the cars are off in all directions, each vehicle to its assignment. For several of the probes, that means traversing the dryline on country roads along the Texas-Oklahoma border.

Storm chasing has become big business, spurred on by the 1996 movie Twister, in which Helen Hunt and Bill Paxton openly flaunt crazed do-or-die attitudes in the midst of snarling black-fingered beasts. Outfitters with names like Violent Skies Inc., Tornado Alley Safaris, and Cloud 9 Tornado Chasing Tours load up their vans with tourists at a going rate of up to $1,500 per person per week, not including airfare or food. Many of the trip leaders are amateurs who have been chasing storms all their lives. Others have meteorology degrees.

IHOP is not about racing after twisters. Rather than focusing on the wrath of a full-blown tornado, the IHOP researchers are interested in tracking the birth of severe storms—the ones that produce lightning, hail, and enough rainfall to cause flash floods, and sometimes tornadoes. They want to understand how storms form, particularly the phenomena that occur on a scale too small to be detected by National Weather Service weather stations. The National Weather Service uses a network of Doppler radars to “see” incoming storms, but because the radar units are separated by more than 100 miles, they usually miss what is happening at boundaries that occur over just a couple hundred yards. The radars don't have the resolution to detect small-scale processes along drylines that form more than a few miles away. “That's why our mobile fleet is so important,” Markowski says.

By venturing into the field, the researchers hope to take snapshots of the atmosphere while storm cells—organized patterns of warm, moist, rising air and cooler, drier, sinking air—are developing. They'd like to figure out why an atmosphere that looks ripe for a thunderstorm sometimes won't produce one, or why a storm may yield localized street flooding when areas only a few miles away have no rain at all.

“Once we get the small picture, we need to pay attention to how those small-scale weather features are linked to large-scale features that we can observe using the National Weather Service network,” Markowski explains. “That way, we can use the large-scale information to infer what's going on at the smaller scale.” Understanding the small picture could lead to improved forecasting of flash floods, which wreck more lives and property than any other type of severe storm.

raised truck with radar in front of stormy skyJames Marquis

A DOW Doppler on Wheels) radar scans the skies as a storm forms. Penn State meteorologist Yvette Richardson and Josh Wurman of the University of Oklahoma (waving from truck) coordinated the movement of radar trucks for the IHOP field experiment.

It was damage caused by tornadoes, so violent and mysterious, that first drew Markowski to the weather. On the evening of May 31, 1985, ten-year-old Markowski watched tornado warnings for parts of western Pennsylvania—not far from his home in Camp Hill—scroll across the bottom of the television screen, interrupting a prime time showing of Superman 2. His parents were “clueless about what that meant,” Markowski recalls. Later, he learned that supercell thunderstorms—storm cells with rotating winds—had crossed eastern Ohio, western Pennsylvania, New York, and Ontario, Canada, spawning over 43 tornadoes. The outbreak lasted for six hours. When it was over, 75 people had lost their lives and damage was estimated at $450 million. Markowski, fascinated by an act of nature he didn't understand, asked his parents to drop him off at the library several times a week that summer so that he could read more about tornadoes and the atmosphere. “I was hooked,” he says. His boyhood passion became his career.

As an undergraduate meteorology student at Penn State in 1995, Markowski spent ten weeks tracking tornadoes in Oklahoma as part of a National Science Foundation Research Experience for Undergraduates program. The project, led by Rasmussen, was called VORTEX, for Verification of the Origins of Rotation in Tornadoes Experiment. The data that Rasmussen, Markowski, and other researchers collected in the field, some of which Markowski later analyzed as a graduate student at the University of Oklahoma, led directly to the formation of several hypotheses that are being tested during IHOP. “During VORTEX, we realized that there was a broader problem: whether or not storms initiated in the first place,” says Markowski. The amount and distribution of water vapor in the air plays a huge role in when and where storms form; yet meteorologists have a very poor understanding of it.

“The meteorology community saw water vapor as something we need to understand better, especially to improve predictions of rainfall,” says Markowski. “There was a big push for this project.” As several researchers note during the long days in the field, IHOP is really about chasing water vapor, not tornadoes.

Chasing water vapor is about as far from Hollywood-style storm chasing as Oklahoma is from California.

It's 3:45 on a hot, dusty afternoon and Probe 8 has been cruising back and forth along the same ten-mile stretch of road for more than two hours. A few puffy clouds have formed in the bright blue sky, but there's no sign of a storm brewing.

Scott Axelson, a recent graduate of the University of Oklahoma, drives with one arm resting on top of the wheel. Andrew Philpott, an undergraduate from Bates College in Maine, mans the laptop computer that collects and displays the data gathered by the instruments on top of the car. Ben Sipprell, a Penn State undergraduate, sits in the back, taking notes and handling radio communications. The students are here as part of an undergraduate research program similar to the one Markowski joined in 1995.

Their mission is to collect data such as dew point—a measure of moisture in the air—and temperature on both the moist and the dry sides of the dryline. Part of the trick is to find the boundary. Change in dew point is the key. Philpott tracks the graphs on the laptop and tells us when the dew point starts to rise, first slowly and then in small jumps. Sipprell explains that the change in moisture over a certain distance is called a moisture gradient, and that the dryline is somewhere on that gradient. As that distance gets shorter, meteorologists like to say that the moisture gradient is tightening. Sometimes, this tightening will result in a storm. Not always, though. The researchers hope that the data will reveal why the process seems to be hit or miss.

dark cloud with a hint of orange far awayJames Marquis

Once a storm initiates, the window of opportunity for data collection has closed. Above, a storm front passes through an area where one of the radar trucks had been scanning.

While the cars are on the ground gathering data, and the planes—also equipped with radar—are in the air traversing the boundary 1,000 feet above the ground, the radars scan the entire field from the four corners of the imaginary box. Three of the radars are called Doppler On Wheels, DOW for short. Developed in 1995 by Josh Wurman of the University of Oklahoma, the DOWs, like all radars, send out a microwave beam to gather data on reflectivity and velocity. When the beam hits a target—rain, for example—the energy scatters and reflects, revealing the size or density of the target. Heavy rain scatters more energy than light rain. The radars also reveal the speed at which the target is moving, such as rain blown by wind. Because the DOWs are fully mobile, researchers can redeploy several times in an afternoon and can even gather data on the road. And while one radar scanning the field reveals only the component of motion toward or away from the radar, two radars reveal the complete wind field. Scans from several radars allow researchers to construct a three-dimensional wind field of a particular area—a pretty green picture of rain aloft in rotating winds, for example. Yvette Richardson and Wurman are in charge of the DOWs; they decide, along with the FC, where each of the radars should go, when they should redeploy, and how they should be scanning. The fourth mobile radar in the field, called SMART, for Shared Mobile Atmospheric Research and Teaching, is operated by Jerry Guynes, an electrical engineer from Texas A&M.

Voices crackle over the radio. “The dry-line has moved. Should we head east to hit it?” asks the team in Probe 6. A plaintive voice from another probe asks to stop for gas and more snacks—fuel for the team. Rasmussen, the consummate radio cheerleader, calls out encouragement: “Hang in there, guys. We're getting good data.”

At 7:00 p.m., Rasmussen calls it quits, and everyone heads back to the Super 8 Motel in Woodward about 120 miles to the east, home base for the next couple of days. Even though they didn't see any storms, Rasmussen seems thrilled at the day's run and praises the probe teams for doing an excellent job. The probes transected the boundary at several different points, from both the dry and the moist sides, gathering data throughout the afternoon as the dryline tightened. Rasmussen could see it all unfold on his computer screen in his field coordinator van. “This is some of the best data collection I've ever seen,” Rasmussen says during dinner at a local bar and grill. “By undergraduates!”

“That's because they're not storm chasers,” says Christina Hannon, a graduate student from Penn State. “They're not hellbent on driving off on their own to find tornadoes.”

Rasmussen nods. “I actually like this better than chasing tornadoes, by quite a bit,” he says.

Hannon rolls her eyes. “Right. And how many tornadoes have you seen?”

Rasmussen grins. “Too many to count. After a while you've seen everything, and it's not exciting anymore.” His grin widens. “Boring, actually.”

During the day's mission, a graduate student in one of the probes had had a fit of temper. Evidently misinformed about the team's objective, he couldn't understand why they weren't gunning it across the panhandle in hot pursuit of tornadoes that might have formed several counties away.

dark sky hangs over orange plainsNOAA

A severe storm rolls across the Great Plains. While this region of the United States is known as tornado alley, flash floods cause more damage here than any other severe storm effect. The data collected during IHOP could lead to improved storm forecasting.

The next morning brings gray skies and light drizzle. The students mill around the parking lot of the Super 8 while Rasmussen and Ziegler agonize over the latest weather information. Sipprell points to the white van at the far end of the lot, the glow of a computer monitor visible through its back window. “During the summer, Erik and Conrad live in that van,” Sipprell says, shaking his head.

The buzz among the students is that they're going to call it quits before the day has even begun. But the sky looks stormy. Long flat clouds stretch high above and far into the distance. Markowski explains that it's an anvil. He grabs a notebook and draws a tall column of clouds over half the page.

“The defining characteristic of a thunderstorm is an updraft of warm air, moving at tens of miles an hour, which is pretty fast compared to the vertical speed of the atmosphere,” he explains. The clouds in the updraft—warmer than the surrounding atmosphere—suck in air and rise, releasing moisture on the way up. Around 10 miles, the rising air hits a cap of warmer air. “Keep in mind,” Markowski adds, “that warm at this altitude is -40 degrees Fahrenheit.” When the clouds hit this “warm” ceiling, they stop rising and spread out, sometimes as far as 200 miles, into a shape resembling an anvil. All storms form an anvil, and sometimes after a storm it will remain for hours. The sky full of high clouds, and the cool air blowing, are indicators that the storms in this area have come and gone.

The brightly colored DOWS, parked in front of the hotel, stand out against the gray backdrop. Painted with a combination of primary colors, they look more like carnival rides at rest than scientific instruments. Richardson and her students drove the DOWS from Liberal, Kansas, this morning, where they are stationed. As she stands in the parking lot waiting for a signal from FC, Richardson admits that she's never been keen on the “hurry up and wait” cadence of storm chasing. But that hasn't dampened her enthusiasm for fieldwork. Out in the field, in the midst of a brewing storm, Richardson gets to see real-time data streaming in. “Half the time I don't even see the storms because I'm glued to the radar screen,” she says, laughing. She stays “glued” to ensure that the data are being collected properly.

As an undergraduate at the University of Wisconsin, River Falls, Richardson studied physics and knew little about meteorology as a science. A summer program in atmospheric science at NASA Goddard sparked her interest. Later, as a graduate student at the University of Oklahoma, Richardson went on her share of tornado chases. “You either become a storm chaser or you're out of the culture,” she says. In 1991, she saw a substantial tornado, and for her Ph.D. work, she decided to focus on severe storms.

Like many atmospheric scientists, Richardson divides her research equally between fieldwork and computer modeling. Sometimes the two parts work hand in hand. Other times they are independent. “Most of what we know about how storms form is based on theory, or numerical modeling,” Richardson explains. “The models can give you clues about what to observe in the field.” But it doesn't make sense to run models at a higher resolution than what you can observe, so gathering real world data helps researchers like Richardson develop more detailed models.

An hour later, Rasmussen announces that the boundaries that might form this afternoon are too far away and not promising enough to make the journey. The armada travels back to Norman to regroup.

That night at a Mexican restaurant, Markowski and his students grumble about the day being a bust, “which in itself isn't so bad,” says Markowski. “The real killer,” he adds, as he playfully thrusts a butter knife in the direction of his heart, is that later in the afternoon storms formed in the same county they'd abandoned in the morning. Markowski and many of the students had, out of habit, turned on the Weather Channel upon their return, only to see tornado warnings for areas near Woodward.

Later Markowski and his students learned that Richardson's team in the DOWs had stayed in the Woodward area and were able to deploy and collect data on a supercell that developed into a tornado.

It's Friday at the deserted municipal airport in Shamrock, Texas. The SMART radar truck, its 8-foot-diameter radar dish mounted on a pedestal attached to the truck's trailer, is pointed northwest, sampling a 140-degree swath.

black sky with purple lighteningJames Marquis

Above, a time-lapse photo of cloud-to-ground lightning across the sky. For severe-storm researchers, returning from the field during a storm is often the most dangerous part of the job.

On the monitor, neat green lines show where air streams form a triple point: Hot dry air from the west joins warm, moister air from the south, and cooler air from the north. These lines are actually formed by great hosts of insects, buoyed by converging air streams and recorded by the radar as areas of high reflectivity. The bugs reveal the location and movement of the fronts.

The SMART radar sits at the southeast corner of an imaginary box formed with the three DOWs. Rasmussen is guessing that the converging fronts will pass right through the approximately 150 square miles of territory he's staked out. By late afternoon, as the ground temperature rises and anticipation grows, the researchers are hoping to see some “CI,” as they call it. Convective initiation. Clouds billowing upward, lightning, precipitation. The works. “So far, we've racked up the ‘no initiation' cases,” says Markowski, who is traveling with the SMART radar team today. “It's time for a storm.”

Collecting meaningful data will be a big challenge. They're trying to trap a moving target. Will the clouds burst in the box or two miles away? The whole armada may have to redeploy several times throughout the day, and while it's easy enough for the geek mobiles and the DOWs to change position, it's a little more difficult for the SMART radar. It takes about half an hour for the SMART truck to set up and start scanning. Steel plates must be placed on the ground so that the trailer's extendable feet don't sink into mud. The dish must be raised. Scanning coordinates must be entered into the computer.

The SMART radar, designed and built by Guynes, the electrical engineer from Texas A&M, is incredibly sturdy; it survived Tropical Storm Gabrielle near Venice, Florida, in September 2001. However, Guynes has yet to construct a protective cover for the dish, making it vulnerable to the golf-ball-sized (and larger) hail that storms on the Great Plains can produce. So Guynes, who is here to operate SMART radar for the IHOP team, warns Markowski that he wants to be on the road heading to safety before the weather gets serious.

As Guynes sits in a lawn chair in a field near the SMART radar truck, Markowski is watching the sky and pacing. It's late afternoon and he's sweating in the hot sun. The clouds are gathering.

During waiting periods like this one, Markowski, who plays part-time with a semiprofessional baseball team, usually pulls out a couple of gloves and a baseball and tries to coax someone to play catch. Today, no one has time. Rasmussen sends a message over the radio to change position. The cold front is moving rapidly to the east and they need to move the sampling box. It takes 20 minutes to pack up and move to the new coordinates about a mile down the road. As soon as they start setting up in the new position they get another call to redeploy. As they pack up for the second time, huge clouds billow upward in the distance.

blue yellow and red satelliteJames Marquis

Driving east, they hit a wall of rain, then hail. Over the radio, Rasmussen suggests that they drive west to get away from it and instead the worst of the storm slams the truck. Marble-sized hail bounces off the windshield. Guynes, concerned about his radar dish, takes refuge under a bridge. Later, some of the students in the geek mobiles say that they saw the clearly defined hail column from miles away. The hail continues for several minutes. After it subsides, Rasmussen must coax Guynes out from under the bridge. The radar dish survives unscathed.

As the armada retreats to Norman, it encounters torrents of rain and blinding cloud-to-ground lightning. The visibility is poor and the students understand now that driving home in a severe thunderstorm is definitely the most dangerous part of this kind of fieldwork.

A couple days later, on the way to another mission on the Great Plains, Markowski explains what happened back in Shamrock. “We were set up in the right spot. The triple point actually moved through the observation box. Initiation occurred just to the east of the box, and the clouds grew to their greatest vertical depth to the east. It was probably a good case, but we won't know until we analyze the data.

“That's the difference between chasing tornadoes and chasing water vapor. You know when you're sampling a tornado because you can see it. But water vapor is invisible. You can't tell if you've captured the beginning of the storm. You won't know until later.”

And later, in the case of IHOP, means several years from now, after hundreds of researchers have culled through and made sense of the hundreds of gigabytes of data that were collected before the storm hit.

Paul Markowski, Ph.D., is assistant professor of meteorology, College of Earth and Mineral Sciences, 503 Walker Bldg., University Park, PA 16802; 814-865-0478; Yvette Richardson, Ph.D., is assistant professor of meteorology, College of Earth and Mineral Sciences, 515 Walker Bldg.; 863-0791; Funding for IHOP 2002 is provided by the National Science Foundation, National Oceanic and Atmospheric Administration, National Aeronautics and Space Administration, Department of Energy, Department of Defense, University of Hohenheim (Germany), National Weather Service, Bundesministerium f r Bildung und Forschung (Germany) and the Deutsche Forshungsgemeinshaft (Germany), CNRS Service d'Aéronomie (France), and Canadian Foundation for Climate and Atmospheric Sciences. For more information, visit

Last Updated January 01, 2003