SkepticblogSkepticblog logo banner

top navigation:

The Tornadoes of 2011

by Donald Prothero, Jun 15 2011

The national news is dominated by yet another set of extraordinary tornadoes in the southern and central United States. The last month brought enormous twisters, including the May 22 tornado that wiped out Joplin, Missouri, and paved a path of destruction in Oklahoma and Kansas as well. It has killed at least 144 people (so far), making it the deadliest single tornado since the April 9, 1947, event that killed 181 in Woodward, Oklahoma. Back on April 27, Tuscaloosa, Alabama, was devastated, with a death toll that is still unknown as searchers comb through the debris. But on that date alone, over 327 tornadoes were reported, causing at least 344 deaths (149 of those in Alabama), with significant damage and deaths from Arkansas to Mississippi and on up into Tennessee and Georgia. The death toll from these storms exceeds the more than 300 killed in the legendary April 3–4, 1974, “Super Outbreak”, which caused death and destruction from Michigan, Indiana, and Ohio to Alabama and Georgia. These recent storms follow closely on the heels of major tornado outbreaks all over the Midwest and Southeast in February, March, and early April.

What is happening? We understand the fundamentals of tornadoes pretty well. Usually there is warm moist air mass rising from the Gulf of Mexico that moves north and meets cooler, drier air from the northern Plains and the Rockies. When these collide, a strong front develops which causes a big horizontal cylindrical vortex to form. The warm air rises up as it meets the cold air and thunderheads grow. If there is also strong shear from the jet stream, the horizontal cylindrical spiral of air will tilt into a vertical funnel. If it continues to grow, it will touch the ground and become a tornado.

Because tornadoes are generated when these different air masses come into collide, they are most common in the spring, when the weather is transitioning from cold on the northern Plains to hot on the Gulf Coast. Thus, March through July are the busiest months of “tornado season”, although tornadoes occur in every month of the year. The U.S. has by far most of the world’s tornadoes due to its favorable geography. The Rocky Mountains funnel and block air masses, the warm Gulf air rises up from the south, and cold air masses descend from Canada.

The only other country in the world with significant tornadoes? Bangladesh, which also has a barrier of high mountains funneling the air masses (the Himalayas), warm moist air from the Bay of Bengal, and other similar factors. The world’s deadliest tornado occurred not in the U.S., but in Bangladesh on April 26, 1989, killing more that 1300 people.

The U.S. tornado season is getting off to a rip-roaring start this year, with over 1208 tornadoes reported so far, at least 875 of those confirmed (as of May 24, 2011). At this pace, 2011 will easily break the previous records for tornado seasons (although there have been seasons in the past that started strong and then fizzled). Of these, there were at least 4 tornadoes that were a Category 5 on the Enhanced Fujita scale, the largest known events, with estimated wind speeds in excess of 200 mph (322 km/hr). So far, 2011 is the deadliest year in U.S. tornado history due to the more than 322 deaths in the April 27 outbreak and the more than 144 deaths in the May 22 outbreak.

The good news is that these numbers don’t yet approach the deadliest tornadoes in U.S. history, including over 700 killed and 2027 injured by the Tri-State Tornado of March 18, 1925. That twister carved a path of destruction for hundreds of miles from Missouri to Indiana. If you look at a list of the Top Ten Killer Tornadoes in the U.S., most occurred in the late nineteenth-early twentieth centuries, when there were no tornado warning systems in place and fewer places had tornado shelters. Today, there may be lots of dangerous tornadoes in a given year, but a higher percentage of people survive because of warnings and shelters and better building construction—even though our population growth and development are putting more people in harm’s way.

Are tornadoes becoming more frequent than ever before, just like larger killer hurricanes are becoming more frequent and energetic? Here, the evidence is still preliminary and inconclusive. Tornado data from the National Oceanographic and Atmospheric Administration (NOAA) seems to show a fairly stable but fluctuating trend of about 800-1000 tornadoes per year between 1970 and 1988. Then, starting in 1989, the trend climbs to around 1200 tornadoes per year, with big spike of over 1300-1400 tornadoes in 1998 and 2003, 1692 in 2008, 1156 in 2009, 1282 in 2010, and already over 1200 in less than 5 months of the start of 2011. There seems to be no strong correlation with El Niño or La Niña years, as once thought. A lot of scientists are currently studying the data, but so far as I know, no one has done a careful statistical analysis to see if this trend is meaningful, and whether it holds up against the records over the past century. There are numerous confounding factors, such as the effects of larger more widespread populations that increase the potential damaged areas, and also a larger, more alert population that reports more tornadoes than 50 years ago. But it would not surprise me to find that the warming of the tropics that drives hurricanes also means more energy in those Gulf Coast warm moist air masses that cause tornadoes, as was predicted in 2007 studies by NASA and the National Academy of Sciences, and the 2008 study “Scientific Assessment of the Effects of Global Change on the United States.”

So hold on to your hats! It may not be the end of the world but by natural variation coupled to a warming trend we might just get walloped more this year than most.

27 Responses to “The Tornadoes of 2011”

  1. Trimegistus says:

    Evidence is “preliminary and inconclusive” — which presumably means “there’s no proof at all yet but we’re hoping!”

  2. Shane Brady says:

    There are so many storm chasers watching almost every cell out there, that even quick touchdowns are getting counted. We really don’t know how many tornadoes there were 50-60 years ago.

  3. Shane Brady says:

    From the tornado data article mentioned:

    “The increase in reported tornado frequency during the early 1990s corresponds to the operational implementation of Doppler weather radars. ”

    “Dr. Changnon has long advocated the use of “event
    days” because of its mitigation of the impact of reporting
    biases (Changnon and Schnickedanz, 1969). When
    tornado days are plotted against year (Fig. 2), the rapid
    inflation that is apparent in the numbers of reported
    tornadoes is no longer present. Instead, only small
    changes are seen over the period 1970 through 2002; “

  4. Somite says:

    Seems like Donald activated the Tornado Misinformation Response Team.

  5. MadScientist says:

    I wouldn’t include any mention of the warming trend since no one has established that global warming = bigger or more numerous storms. Intuition is not science and I’ve worked on enough projects where my calculations have run counter to what people would expect (but were correct) that I no longer give any credence to gut feelings. The claim on the IPCC web site (at least when I looked a few years ago) that it is “difficult if not impossible” to establish that warming = bigger or more numerous storms is not a scientific claim.

    • tmac57 says:

      It’s an interesting question to ponder.At what point can we conclude that climate change is driving the more extreme weather events? The pattern so far is inconclusive,but if sometime down the road there were twice or three times as many events in multiple categories , that tracked steadily with increased temperature,wouldn’t that reduce uncertainty to the point that one could practically say there was causation? Where do we set the bar?

      • MadScientist says:

        Sure, I’m not opposed to continuing to analyze data to see if warmer = more numerous and more violent storms. I only object to people in the present (and in the past) stating that hypothesis as a given fact while in reality having no data to support their assertions. It’s nice to see that some people have already done the work on US tornadoes. In the future the challenge will be to compare the developing reporting techniques with the current reporting techniques in order to make the figures more easily comparable and hopefully identify any correlation with the global temperature trend.

      • Max says:

        If the temperature keeps increasing, wouldn’t any similarly increasing or decreasing trend correlate with it, such as the number of pirates in the world?
        (That graph is outdated)

      • MadScientist says:

        Sure, but there is no sensible hypothesized mechanism by which pirates make a significant contribution to the strength or number of storms. At the moment the challenge is to show even so much as a correlation between warming and number or intensity of storms – once a correlation is established then people can look at it and have serious discussions over whether or not there is a causal link.

  6. CountryGirl says:

    It is a mistake bordering on intentional misrepresentation to cite deaths from tornados as “proof” that things are getting worse. Part of the reason is our population is growing so every year more people are likely to be affected by a natural phenomenon. In 1974 the U.S. population was 2/3rds what it is today. The population in most cities and urban areas today have grown faster then the population of the country as more people move to cities. In 1974 in addition to the terrible tornadoes that destroyed places like Xenia Oh. also tore up huge expanses of farmland killing only cattle or corn plants. Today those same tornadoes would be destroying homes where those cornfields once stood.

    • Max says:

      Dr. Prothero addressed this:
      “Today, there may be lots of dangerous tornadoes in a given year, but a higher percentage of people survive because of warnings and shelters and better building construction—even though our population growth and development are putting more people in harm’s way.”

    • Hah … foiled again, this time by Max.

  7. Max says:

    How does the death toll from tornadoes compare with the death toll from heat waves so far?

    • As I explained in my new book “Catastrophes!”, tornadoes are pikers compared to heat waves. Even this bad year killed only a few hundred, whereas some of the bad heat waves kill thousands to tens of thousands. That’s the general point I brought up in my earlier blogpost: the deadliest disasters are not the scary ones like earthquakes and tornadoes, but the subtle everyday ones: heat waves and blizzards.

      • WScott says:

        Since I happen to have the numbers handy: 55,736 people died in Russia’s 2010 heat wave, making it the second deadliest disaster of last year. (Behind the Haitian earthquake, which killed 316,000+.) And that’s just in one country.

  8. Trimegistus says:

    Cold weather = “just anecdotal, not data.”
    Storms making the news = “PROOF!”

    • No, that’s not the claim of AGW models at all. They all predict warmer temps = more energy and moisture in atmosphere = more frequent and severe storms, both snowstorms and also hurricanes and tornadoes. Local weather is indeed unpredictable, but long term increases in frequency and/or energy of storms IS a prediction of the models, including the three I just cited at the end.

    • MadScientist says:

      That’s funny, I thought cold weather = “Proof!” and global mean temperature over decades = “Conspiracy! nanananananana I can’t hear you! nanananananana!”.

  9. Joe says:

    Actually, if tornados are caused by a collision of a warm air mass with a cold air mass and there are no cold air masses (global warming) wouldn’t there be fewer and less powerful tornado’s?

    • Donald Prothero says:

      No, even in a warmer earth there is always less sunlight on the poles than on the equator, so there is always a latitudinal gradient in temperature and energy received. From what I’ve heard, the crucial factor is how much warmer the equatorial waters of the Atlantic and Pacific have become in recent years, spawning more frequent and more energetic hurricanes and typhoons. And during the winter time, the poles still freeze. In the springtime, those cold air masses meet warmer than usual masses in the Tornado belt, which is why those reports I cited are predicting more frequent and more powerful tornadoes than previously.

  10. Nestor says:

    “From what I’ve heard, the crucial factor is how much warmer the equatorial waters of the Atlantic and Pacific ”

    This of course is counter to the best data collection system available and tends to exist only in “models”

    Argo shows no rise in sea temp.

    • tmac57 says:

      From Argo’s own website:

      “The global Argo dataset is not yet long enough to observe global change signals. Seasonal and interannual variability dominate the present 6-year globally-averaged time series. Sparse global sampling during 2004-2005 can lead to substantial differences in statistical analyses of ocean temperature and trend (or steric sea level and its trend, e.g. Leuliette and Miller, 2009). Analyses of decadal changes presently focus on comparison of Argo to sparse and sometimes inaccurate historical data. Argo’s greatest contributions to observing the global oceans are still in the future, but its global span is clearly transforming the capability to observe climate-related changes. “