A tornadic thunderstorm approaches South Haven, Kansas on May 19, 2013. (Reuters)
"You had to be underground," The Weather Channel's Mike Bettes put it, "in order to survive this tornado."
He was not exaggerating. The massive storm that hit Moore, Oklahoma yesterday -- featuring winds up to 200mph -- didn't simply put those aboveground at risk of being swept up in its funnel. The tornado also took the smallest objects of everyday life -- down to pebbles and even dust -- and effectively converted them into bullets and shrapnel. Much of the carnage that so often results from tornadoes is the result of a terrible phenomenon known by an appropriately terrible euphemism: "debris impacts."
Larry Tanner is the manager of the Debris Impact Test Facility for the Wind Science and Engineering Research Center at Texas Tech University. In that capacity, he focuses on the (relatively) optimistic side of devastating storms: sheltering people from them. One of Tanner's jobs is to test shelters and their components in his lab, creating several-hundred-mph winds and debris storms -- which can't be mathematically simulated -- to analyze those shelters' ability to withstand a natural storm. Another of his jobs is to work with FEMA to assess the performance of storm shelters after tragedies like the Oklahoma tornadoes of 1999. Or the Missouri tornado of 2011.
Or, now the Oklahoma tornado of 2013.
There are thousands of shelters in Oklahoma -- to the extent, Tanner told me, that the state "has been the most active shelter state, probably, within the United States." Yesterday, however, that wasn't enough. The storm was deadly in part because there were so few places -- underground places -- for people to escape to. As Weather Nation's Paul Douglas noted last night, fewer than one in 10 Oklahomans have access to the basements that stand the best chance of keeping them safe when a "monster" -- another appropriately awful euphemism -- strikes.
So why weren't there more underground shelters for people to escape to yesterday? Why aren't houses in the vast stretch of prairie known as Tornado Alley routinely equipped with storm cellars?
There are, unsurprisingly, several reasons. But what they come down to is this: Yesterday's tornado was deadly, ultimately, not just because of the vagaries of the sky. It was deadly, too, because of the fickleness of the ground.
Ground That's Made of Clay
The relative dearth of storm cellars in Oklahoma may be partially attributed, as things so often can, to environmental factors. The soil in the state is composed largely of clay -- and that's particularly true in central Oklahoma, where Moore is located. ("Soils in the Central Rolling Red Prairies," geologists at Oklahoma State put it (pdf), "are dark and loamy with clayey to loamy subsoils developed on Permian shales, mudstones, sandstones and/or alluvial deposits under tall grasses.")
The ground in central Oklahoma tends to be soft and moist -- right down to the bedrock that sits, generally, some 20 to 100 feet below the surface.
A soil map of Oklahoma: Moore is located in Oklahoma County, in the loamy-soiled center of state. ( Brookhaven National Laboratory)
Here's the problem with that when it comes to building basements and underground shelters: Clay is particularly fickle as a foundation for construction. When loamy soils absorb rainwater, they expand. And when the weather's dry, they contract. This inevitable and yet largely unpredictable variability makes basement-building a particular challenge, since it makes it nearly impossible to establish firm foundations for underground construction.
And while above-ground homes can be built on these somewhat shaky foundations, adding the element of open space in the form of a basement is a nearly impossible feat of engineering. There is a chance your house, its basement surrounded by glorified mud, will eventually simply topple into itself.
To mitigate this, contractors have been experimenting with steel reinforcements for basements, bolstering underground walls with steel beams that are drilled directly into the bedrock below. The problem here, though, is that much of Oklahoma's bedrock is composed of limestone (pdf), which, just like the soil above it, absorbs water. And which, when it's sapped of moisture, becomes chalky.
So, you can try to enforce your basement with steel; ultimately, though, the steel will be anchored to rock that is "rock" only in the broadest sense of the word.
'Bill Gates's Wealth'
There's an obvious corollary to these challenges. If you're going to attempt to build a steel-reinforced shelter -- the only kind that, despite its drawbacks, would stand a good chance of keeping you safe during a storm of yesterday's magnitude -- it's going to be expensive. Underground shelters cost thousands of dollars, at minimum, and many of them much more than that. Which are not sums that many families, especially given the current economic conditions, have at their disposal. As Paul Douglas summed it up: For most people, "it's cost-prohibitive to dig basements."
Part of the expense comes from the fact that guaranteeing a storm-resistant basement -- ensuring that your cellar is also a storm cellar -- requires more than just an underground space. As the Texas Tech's National Wind Institute explains:
The reality is that the walls, roof, windows, doors, and garage doors must be missile-resistant and the connections of the structural elements must be capable of withstanding 250 mph wind pressures. Missile resistance of walls and even roofs is fairly easy to achieve with current ICF (insulating concrete form) construction, however the connections for long span roofs and tall walls to transfer the loads induced by a 250 mph wind must be 7 ½ times stronger than those routinely required by today's codes. Then, the issue is how to protect the numerous openings (doors and windows) routinely found in a home.... Research has shown that what is practical and reasonable is to do a better job of building your home with good wind resistance connections (wind clips, anchor bolts and proper nailing of wall and roof diaphragms) and include a shelter within the home.
Yes. And none of that comes cheap. To the question of whether it's even possible to build a tornado-proof house, the Wind Institute offered the following reply: "You can, but your neighbors probably would not like it in their neighborhood and you would need some of Bill Gates's wealth to pay for it."
Safety in Numbers (and the Home Depot)
Which means that many people, in the event of a major storm, rely on community shelters. And that's a relatively recent thing. Six months before the storm that devastated Moore in 1999, FEMA published P-320: a guide to "building a safe room for your home or small business." (Tanner was one of the authors of that document.) The guide included detailed construction plans and specs for the DIY construction of safe rooms. And it was telling: At the time, Tanner notes, "We had no design guide for community shelters in terms of schools and commercial buildings and federal buildings."
Then came the storm, and with it the tornado that is the most powerful the country has ever seen. As the community prepared to rebuild itself, President Clinton declared that the residences, schools, commercial areas, and federal buidlings being reconstructed after the storm should include safe rooms in their layouts. The state of Oklahoma used a substantial portion of the mitigation funding provided by FEMA to supplement the construction of storm shelters in Oklahoma. "And that," Tanner says, is what "really spawned the safe room and the below-ground shelter industry."
Personal, family-oriented shelters are still the most common kind, Tanner says. But the impact of "May 3," as it's often shorthanded, also emphasized the role that communities would come to play in offering shelter to their residents. The old, Wizard of Oz-style model of sheltering -- every farm with its cellar -- is slowly giving way, in the age of suburban sprawl, to large shelters meant to house large groups of people. Following May 3, Tanner and his co-authors at FEMA and several universities scrambled to write a new document, P-361: a design guide, this time, for community shelters.
This safe room, at Oklahoma State's Child Development Center Lab School in Oklahoma City, doubles as an exercise studio and play area. ( OSU-OKC)
And those large shelters, Tanner says, are now becoming more common in places like mobile home parks and schools -- areas that house large concentrations of people over long stretches of time. Commercial centers are making efforts in that regard, as well: After the Joplin, Missouri tornado of 2011, a Home Depot that had been destroyed in the storm was rebuilt with a concrete-reinforced room -- capable of holding150 people -- included in its layout.
Sheltering, in other words, is moving from an individual concern to a collective one. And while community shelters still aren't the norm -- Moore, for example, has no official public shelter, on the logic that sheltering-in-place is often the most prudent strategy -- that shift makes sense. More public shelters will mean more options for people when tornadoes form, as they do, with little notice. One reason tornadoes prove so deadly now is that, given the spread of the suburbs, their funnels simply stand a better chance of touching down where people are. Tornadoes used to hit farmland; now, that same land is increasingly concentrated with people. If the suburbs keep expanding, Paul Douglas put it, "we're going to see more of this."
The question is how well-prepared we'll be when we do.
More From The Atlantic