By Jim Schutze
By Rachel Watts
By Lauren Drewes Daniels
By Anna Merlan
By Lee Escobedo
By Eric Nicholson
Marshall says he is accustomed to taking heat for his views, which become relevant during insurance disputes and court cases with large damage awards at stake. Over the years, he has been knocked as both a shill for insurance companies—for suggesting that water did most of the damage along the Gulf Coast in Hurricane Katrina—and an enemy of the free market. The later was pinned on him when he advocated requiring mobile home parks in Kansas to provide residents with tornado shelters. The mobile home lobby knocked the idea flat, he says.
For a house hit by an F5 tornado working at full force, even well-fastened anchor bolts wouldn't make a difference. But almost all tornado damage occurs at lesser wind speeds, Marshall says.
Experts spent considerable time debating the strength of the May 27, 1997, tornado that swept through Jarrell, north of Austin, and killed 27 people. Storm-damage photos showing clean-swept foundations where 40 single-family homes once stood lent support to the National Weather Service's official finding that the tornado reached F5 intensity.
But in a report the following year, a team from the federal Building and Fire Research Laboratory concluded that the twister was a tamer F3. Homes in the subdivision, which was located on county land and not regulated by building codes, were poorly constructed, fastened down with nails and likely were knocked down and carried away by lesser winds, the report found.
Government safety warnings, developed from experiences in past tornadoes, instruct people to take shelter in a central, ground-floor room of their house rather than attempt to flee. The risks of taking flight on the roads were borne out in the April 10, 1979, Wichita Falls tornado. Twenty seven of the 42 people killed were attempting to escape the mile-wide wedge funnel in their vehicles.
Advice to stay put worked poorly in Jarrell, but extensive reviews of the Oklahoma F5 suggest it served people well there. Emergency officials in Moore estimate that 2,000 people were in the twister's path and took shelter, like the Bernich family, by huddling low in their house's or apartment's interior rooms. Just three were killed, an outcome federal officials describe as remarkable given the tornado's intensity.
Photos and video footage of the damage show trees stripped to their trunks and houses reduced to lumber piles. But a closer look shows many homes survived with a few interior walls standing—just enough space to furnish shelter. "We found that the majority of people who sought shelter inside their homes survived the tornado without serious injury," says Marshall, who worked with a Texas Tech team surveying the damage. In a report on its findings published in 2002 in the journal "Weather and Forecasting," Marshall wrote that in developed areas, tornado damage begets more tornado damage. "As homes broke apart, their debris impacted neighboring homes, compromising them as well."
Except in rare instances, such as a subdivision that was flattened less than a mile east of where he videotaped the twister, better-built homes survived.
Still, a tornado loose in Dallas with the power of the one he witnessed less than 200 miles from here would be, in his words, "a disaster of unprecedented proportions." And it could happen, adds Marshall. "I hate to Mr. Doom and Gloom about it, but every year without a major strike here just adds to the apathy that is out there."