It struck almost without warning nearly 10 years ago today.
On June 19, 1981, a tornado carved a path of destruction through Lawrence. One man died and 33 other people were injured. Damage was estimated at about $18 million.
"I think about it almost every time I look out of my right eye," said Phil Rankin. Rankin was a volunteer storm spotter who was hit with debris when he was caught directly in the path of the storm near 27th Street and Lawrence Avenue.
The tornado, which dropped out of the sky shortly after 7:30 p.m., skipped along a two-mile path between 27th and 31st streets in southwest Lawrence, demolishing the south end of the K mart store, 3106 Iowa, and overturning mobile homes at Gaslight Village mobile home park, 1900 W. 31st.
STAN PITTMAN, a 30-year-old Kansas University computer assistant, was killed by falling debris inside the K mart store.
The tornado also damaged or destroyed at least 15 homes and several other businesses.
"People didn't have time. They didn't have much warning," said Rankin, who, despite having glass fragments propelled into his body by the storm's winds, was able to issue a warning through his radio that a tornado was on the ground.
Rankin's words started the city's warning sirens, but residents only had about 30 seconds to take cover, he said.
Although the 1981 storm changed many residents' lives, Rankin said the destruction could have been much worse.
"We were very lucky that we didn't have an Andover (disaster) on our hands," he said. Andover, a small town near Wichita, was devastated by a tornado this spring.
THE TORNADO that struck Lawrence that day was unusual, Rankin and meteorologists say.
It was imbedded in a thunderstorm that intensified so rapidly that forecasters at the National Weather Service did not have time to warn local residents.
Rankin said the thunderstorm that spawned the tornado grew from a height of 38,000 feet to 62,000 feet in 10 minutes.
And, perhaps most significantly, few residents actually saw a funnel cloud with the killer storm.
"THERE ARE no statistics, but probably less than 25 percent of all tornadoes have a vortex that you can't see," said Joe Eagleman, Kansas University professor of meteorology, physics and astronomy.
"There are very few reports of anyone seeing a funnel with the 1981 tornado," he said.
Eagleman, who watched the 1981 storm from Pleasant Grove Hill, just south of Lawrence U.S. Highway 59, defended forecasters, who only had issued a severe thunderstorm watch for the area earlier in the day.
Eagleman was quoted shortly after the storm as saying, "We expect quite a lot from our forecasters, but the state of the art is such that fewer than 50 percent of the big storms that develop suddenly and hit like this one did are ever forecast because there is no way for our weather observers to be everywhere at once . . ."
TODAY, FORECASTERS still cannot be everywhere at once, but plans for a new radar system may allow them to cover tornado-producing thunderstorms as never before.
"It will be a totally radical change from what we now have," said Philip Bills, lead forecaster with the Topeka weather service office.
Bills and other meteorologists are excited about NEXRAD, or Next Generation Radar, a radar system the will vastly improve the data on thunderstorm development and lifespan.
A total of 113 NEXRAD radar systems are scheduled to replace the current radar at weather service offices in the country by the mid-1990s, said Gary Foltz, NEXRAD meteorologist at the weather services' regional office in Kansas City, Mo.
KANSAS CITY is scheduled to have the new radar installed next year, and Topeka should have a system installed by 1993, Foltz said.
The Lawrence area would be covered by radars at both sites, he said.
The NEXRAD system will allow forecasters to observe not only storm precipitation, but will allow them to see the wind direction and speed at virtually every level within a storm.
"Everyone's excited about it," Foltz said. "It just gives you so much more information to work with."
Most current radar systems only show precipitation intensity at a fixed level of a storm within a 125-mile radius of the radar site, he said.
Tornadoes, which are seen as "hook echos" on current radars, only appear if precipitation is circulating around the funnel.
EAGLEMAN SAID only about 30 percent of all tornadoes are detected by the current radar systems.
The new radars, using sophisticated computers and color-coded maps, will provide better range and resolution.
Tornadoes often form in an updraft area in the rear of a thunderstorm, Eagleman said.
He said that new radar system will enable forecasters to monitor where circulaton is present in the storm, and if it is intensifying.
With such information, forecasters could extend the warning time for storms that are likely to produce tornadoes up to 30 minutes, he said.
"The system would enable the forecaster to see a tornado long before it ever reaches the ground," Eagleman said.
THE FIRST NEXRAD system, already operating on Norman, Okla., has been successful in documenting tornadic circulation as far as 200 miles away from the site, forcasters said.
Rankin said weather observers are waiting for the day when Lawrence will be covered by the new systems.
"If you have a wish list as an observer, that is on it," he said. "With the NEXRAD, if we have a tornado like what we had in 1981, we might at least have five minutes (warning) instead of 30 seconds."