Your Odds of Being Hit By a Tornado

Jon Erdman
Published: April 24,2015



 

Have you ever wondered about the odds you'll experience a tornado in your area in any given year? A number of studies over the past few decades have attempted to quantify your tornado risk.

Tracks of reported U.S. tornadoes from 1950-2014.
(NOAA/NWS/SPC)

If you simply examined a map of all U.S. tornado tracks from 1950-2014, you might conclude your odds are disconcertingly high in the nation's mid-section and South.
Certainly, those are the tornado hot zones well known by most. Oklahoma, Arkansas, Kansas, Texas, Alabama and Mississippi may first come to mind.
Approximations of tornado risk have been calculated by a number of studies over the past several decades.
Some of the earliest work to answer this question was conducted by Joseph Schafer and Donald Kelly in the 1980s. A 1986 study concluded that the tornado risk within any 1 degree latitude (about 69 miles wide) by 1 degree longitude (about 53 miles wide in the mid-latitudes) grid box is a maximum over central Oklahoma with about a 0.06 percent yearly risk.
(MAP: Tornado-Free Zip Codes 1950-2013)

Average number of days per year with at least one tornado within 25 miles of a point, based on 1980-1999 data.

Extensive tornado and severe thunderstorm climatology studies were conducted in the early 2000s by Harold Brooks from the National Severe Storms Laboratory, among others.
A 2003 study by Brooks, Charles Doswell and Michael Kay used the concept of a "tornado day," namely, a day in which at least one tornado was reported within 25 miles of any location. This particular 25-mile standard is used in tornado probability forecasts issued by the Storm Prediction Center today.
While not as granular as you may like -- down to your city or street, that is -- a tornado within 25 miles is uncomfortably close, right?
Utilizing 20 years of data from 1980-1999, the Brooks et al. study found a C-shaped area from Illinois to the High Plains south to Texas and east to Mississippi, as well as the Florida Peninsula with at least one tornado day each year. Maxima in tornado days were found in northeast Colorado and parts of central Florida.
Of course, the tornado threat varies during the year, generally speaking, with the return and departure of warm, humid air. Based on the work by Brooks and others, a climatology of seasonal tornado threat was compiled, giving emergency managers and the public an idea when their tornado threat typically peaks.
(MORE: Strange Tornado Damage | Strangest Tornado Locations)
However, the 2003 Brooks et al. study readily acknowledged the limitations of the tornado database at the time, opting to compute tornado days and use only initial touchdown points of tornadoes.
The advance of technology such as Doppler and dual-polarization radar, cellular networks, smartphones, as well as a far greater number of storm spotters and the general increase in population means more weak tornadoes (EF0, EF1) are detected or observed today than, say, in the 1960s. This is known as "tornado inflation."
Secondly, reporting of tornadoes and damage surveys weren't as standardized in decades past as they are today, making tornado tracks of the past more problematic.
Benefiting from over a decade of additional tornado data, including more standard tornado damage surveys, a 2014 study by Timothy Coleman and P. Grady Dixon took another wag at this.
First, Coleman and Dixon used a statistical technique to incorporate entire tornado path lengths, not simply a tornado's start point, to more accurately depict the risk and potential for destruction.

Analysis showing the average annual path length (in kilometers) of all F/EF2 and stronger tornadoes passing within 40 kilometers (25 miles) of a point between 1973 and 2010.
(Coleman and Dixon, 2014)

They also considered only tornadoes of F/EF2 intensity or stronger from 1973-2011, the beginning of which corresponded with the adoption of the Fujita scale by the National Weather Service. Unlike their weaker cousins, these so-called "significant" tornadoes have actually been on a slight diminishing trend over the years, despite better technology and larger storm chaser networks, providing a more stable tornado database not affected by tornado inflation.
(MORE: How Tornadoes Are Rated)
From 2000-2010, F/EF2 or stronger tornadoes accounted for only 10 percent of all U.S. tornadoes, but 92 percent of all U.S. tornado deaths, according to severe weather expert Dr. Greg Forbes.
What they found was a corridor from central Oklahoma into Arkansas, northern Louisiana, much of Mississippi, western and middle Tennessee, and the northern two-thirds of Alabama with the greatest tornado risk, defying the persistent notion of a Plains "Tornado Alley" maximum.
Removing the record-setting April 2011 Superoutbreak had little impact on the spatial distribution.
From 1973-2010, the study found an average tornado path length each year of over 7 kilometers -- which is about 4.3 miles -- within 25 miles of any point in parts of central and eastern Arkansas, central Mississippi and northern Alabama.
For a broad swath of the central Plains, Mississippi Valley and Lower Ohio Valley, Coleman and Dixon found an average yearly tornado path length of at least 3 kilometers -- about 1.8 miles -- within 25 miles of any point.
The probabilities of a tornado actually tracking over your one-half acre lot are of course much lower. However, just knowing you live in an area averaging one tornado day and a tornado path of a few miles nearby each year should make you take every tornado warning seriously.
(MORE: Struck Twice...America's F5 Tornado Towns)
Severe weather expert Dr. Greg Forbes issues TOR:CON tornado threat forecasts every day using a simple scale of 0 to 10, rating the chance of a tornado near your location. Notice a chance of thunderstorms in your forecast? Check the TOR:CON forecast to see if the odds of a tornado in your area are higher.
Being aware of the forecast, knowing where to take shelter, then taking quick action when a warning is issued can stack the odds of surviving a tornado in your favor, even if the odds of experiencing a direct hit are extremely low.

MORE: Metro Area Tornadoes By Month Since 1950