In May, the nation observed the incredible damage that occurs when an F-5 tornado on the Fujita Scale moves across a densely populated area—in this case, the Oklahoma City suburb of Moore. The death toll of 24 was tragic; the advance warning of at least 36 minutes (as much as 42 minutes in east parts of the city) cut the death toll in half.
Considering it only takes seconds to dash down the stairs in a home with a basement, get into a closet or bathtub in a home without a basement, and less than five minutes to move children into the interior of a school building, it is apparent that more warning time, by itself, is not the sole answer to saving additional lives.
There are multiple challenges to building a society that is more resilient to tornadoes and other severe storms. When it comes to thunderstorm-related hazards, an hour or less is the “warning” timescale with longer periods of times considered forecasts.
1. Better Forecasts
If we can, two to three hours in advance, tell people where a tornado is likely to occur, we give those in mobile homes, without basements, and other high-risk situations the opportunity to evacuate the area similar to what is done in hurricanes. This capability does not exist today. It will require computer modeling at approximately four times the scale (i.e., 1 km) than is possible today with faster turnaround. For example, it takes about two hours to make a thunderstorm-scale forecast at 4 km resolution today. That is too slow for the type of forecasts needed.
The U.S. weather satellite program, essential for this type of forecasting to be even possible, is currently in serious trouble. Mismanagement in recent years has left us vulnerable to a satellite failure when we should be leading the world in remote sensing and launching state-of-the-art platforms. For the first time, private sector weather satellite companies are now entering the scene and we may be moving toward private rather than governmental weather satellites.
2. Improved Warnings
We need to get the lead-time (the interval from when the warning is issued to when the storm arrives) for significant tornadoes (those F-2 to F-5 intensity) to a consistent 20 minutes for any given location in the path. After about 20 minutes, there is evidence the benefit of additional lead-time, for the purpose of sheltering, starts to diminish. People will stay in a closet or bathtub for only so long. When there is too much time, there is anecdotal evidence that many stop concentrating on saving their lives and begin to worry about secondary issues (“Should I move things out of the way in the garage to put my car in?” “Should I go up to the attic to get the family album?”). We also know there is a trade-off between lead-time and accuracy. I believe consistent 15- to 20-minute lead-times are achievable in the reasonable future. Striving for significantly longer lead times may increase false alarms that cause the public to lose confidence in the warning system.
What must be done to obtain a consistent 20 minutes of warning?
The radars, which were revolutionary when installed in the early to mid-1990’s, are aging. As far as I can determine, there is no formal plan to replace them, which, given federal procurement cycles, will take at least a decade. The next generation of radars will need to survey the atmosphere more often than the current generation. One approach may be to install a sub-network of less expensive radars that operate on shorter wavelengths (C or X band) to do the more frequent surveying to supplement the existing network. Another proposal is to use phased array radars to track both aircraft and weather (a joint radar for the National Weather Service and Federal Aviation Administration).
We need denser measurements of the upper atmosphere. Traditionally, this is done with weather balloons but those are expensive, labor intensive, and limited to locations where they can be a launched over land which leaves large gaps over oceans. Currently, many jumbo jets are equipped with instrumentation for coverage over land and sea air routes. An issue I see is the proliferation of regional jets. Some of them should be equipped with this instrumentation to improve geographic coverage.
Finally, the traditional two polar orbiting weather satellites that take critical vertical measurements of temperature and humidity should be doubled to four.
3. More Resilient Structures
The cost to make both homes and public buildings more resistant to tornadoes is relatively low. As it is today, the design of some schools enhances the danger rather than mitigates it. For example, a typical design of a school’s administrative area has an unsecured wall up to about seven feet with glass above. Those walls can fall on students sheltering in the hallway. These designs should not be allowed anywhere in the U.S., as they are dangerous in earthquakes as well as tornadoes. In the traditional Tornado Alley and South, the schools should be strengthened against high winds.
4. Finally, Personal Responsibility Plays A Role
Weather science has progressed to the point that all storm warnings should be taken seriously. When the sirens are blaring, the TV and radio are broadcasting tornado warnings and the sky is dark, you are foolish not to take safety precautions. Yes, you might spend a few minutes in shelter and nothing happens. That is a small inconvenience when compared with the risk of losing your life.
These four areas of concentration will likely bring about a further lowering of casualties from tornadoes and other thunderstorm-related causes.
This article is commissioned by Qualcomm Incorporated. The views expressed are the author’s own.