As broken records go, this is a happy one. The United States has now gone an unprecedentedly long time between tornadoes rated EF5, the highest ranking on the Enhanced Fujita Scale, the gold-standard technique for assessing a tornado’s strength based on the damage it wreaks. Unfortunately, as with the stock market, recent performance does not guarantee future behavior.

As of Tuesday, May 25, 2021, it had been eight years and five days since the nation last saw a tornado rated EF5 by the National Weather Service. This broke a record-long quiet spell in a National Weather Service database going back to 1950, taking into account F5 tornadoes on the original Fujita-Pearson Scale as well as EF5s on the Enhanced Fujita Scale that debuted in 2007. (Ratings from the F and EF scales denote comparable damage, but estimated winds are lower on the EF scale.)

The previous record-long gap lasted from May 3, 1999, when a catastrophic F5 tornado slammed Moore, Oklahoma, and Oklahoma City, until May 4, 2007, the date an EF5 destroyed much of Greensburg, Kansas.

The nation’s most recent EF5 ripped across hapless Moore, Oklahoma, on May 20, 2013.

The term “violent tornado” is typically applied by the National Weather Service to the two strongest types, EF4 (top winds of 166-200 mph) or EF5 (greater than 200 mph).

EF5s are infrequent — they account for less than one in a thousand U.S. tornadoes — but they represent some of the nation’s most catastrophic tornadoes on record. The horrors of a top-end tornado were amply demonstrated in Joplin, Missouri, 10 years ago, on May 22, 2011. On a day that didn’t look particularly dire in advance, conditions aligned by evening to generate a large, violent tornado nearly a mile wide that developed just west of Joplin, Missouri, and plowed through the heart of the city.

The rain-wrapped tornado didn’t appear as a classic visible funnel, which gave people scant time to react. What’s more, Joplin hadn’t had recent experience with violent tornadoes, and an unusual sequence of warnings and sirens may have engendered confusion.

The EF5 tornado in Joplin took 158 lives, making it the nation’s deadliest tornado in more than 60 years.

There have been 58 F5 or EF5 twisters since 1950, a little fewer than one per year on average. However, it’s common for several years to go by without any EF5s, followed by several over just a few days’ time, or even on a single day. The Super Outbreaks of 1974 and 2011 produced seven F5s and four EF5s, respectively.

Similar patterns in F5 frequency extend back in unofficial data at least to 1880, according to Tom Grazulis, considered by many to be “the dean” of independent U.S. tornado researchers. The record-long EF5 “drought” in this longer, unofficial database is eight years and 45 days, which would be broken on July 5, 2021, as noted by weather historian Christopher Burt (see Burt’s F5/EF5 analysis in a 2020 Weather Underground post).

Why has there been an EF5 tornado ‘drought’?

One possible explanation for the apparent eight-year absence of EF5 tornadoes: Some tornadoes with EF5-strength winds simply don’t get documented as such, which adds a dose of uncertainty to assessing gaps between EF5s.

The Enhanced Fujita scale is a damage-based tool, relying solely on observed destruction to estimate a tornado’s peak winds. The EF scale uses a far more comprehensive set of damage indicators than did the original F scale. Even so, if an intense tornado plows through a field and hits nothing else, it could end up rated EF0 — even if it had EF5-level winds — simply because there wasn’t a demolished building or other object that could justify a higher rating.

“There’s some monsters out there where you have square miles of nothing but corn,” says Grazulis. “I’m suspecting that some of them are legitimate EF5s but they don’t record as EF5s because there’s not enough that they hit.”

In fact, there may be dozens of U.S. tornadoes each year with top winds in the EF4 and EF5 ranges that aren’t getting rated as such, according to Joshua Wurman, Karen Kosiba, and colleagues at the Center for Severe Weather Research. The group analyzed 120 supercell tornadoes sampled by the center’s Doppler on Wheels mobile radar units between 1995 and 2006 for a paper published in January 2021. A follow-up paper will examine data through 2019 in more detail.

When the team zoomed in on the 82 supercell tornadoes that had official EF ratings and Doppler on Wheels data, they found that the surface winds inferred from Doppler on Wheels were an average of 43 mph stronger than the EF ratings would suggest.

As a result of such mismatches, an expert committee is working to craft a new standard for the American Society of Civil Engineers and American Meteorological Society that would enable analysis and archiving of peak-wind estimates from a range of sources for each tornado. These could potentially include data from radars, anemometers, patterns of tree fall, and forensic engineering techniques. Some wind estimates may be higher or lower than the EF-associated range.

It’s conceivable that some past tornadoes will get reevaluated in light of the new standard, much as scientists have taken a fresh look at historic hurricanes. “At the end of this standards process, I think there’s going to be some serious reanalysis,” says James LaDue (NOAA), who co-chairs the committee with Marc Levitan (National Institute of Standards and Technology).

Is climate change affecting tornadoes?

Climate change does seem to be affecting some aspects of tornado climatology: Twisters appear to be clustering into a fewer number of days with tornadoes, and the areas of peak tornado frequency may be shifting eastward. However, climate change isn’t changing how the geography of the central and eastern U.S. can foster the stew of ingredients required for an EF5 tornado. That includes moist, unstable air from the Gulf of Mexico overtopped by dry, warm air from the desert Southwest and colder air even higher aloft, all flowing at angles that collectively favor storm rotation.

Wurman, who’s observed hurricanes and also tornadoes with Doppler on Wheels, advises taking heed from history. “My personal observation is that it has been a pretty light few years for tornadoes, at least in the areas in which I work or follow. But there have been other similarly slow periods.

“While it may be that tornado frequency, geographical distribution, or intensity are changing due to long-term or short-term climate changes, anecdotal recollections are fraught with the ‘When I was a kid there were blizzards every year’ recollection bias problem,” he says.

Quiet spells can turn on a dime

If there’s one thing to keep in mind about the absence of confirmed EF5 tornadoes over the past eight-plus years, it’s that apparent trends can reverse quickly.

Consider this: It wasn’t so long ago that the U.S. Gulf and Atlantic coasts were enjoying an unprecedented hiatus from landfalling major hurricanes, which are those designated Category 3 or stronger. Going into 2017, there hadn’t been a single Category 3, 4, or 5 landfall since Wilma hit southwest Florida in October 2005. The notion of a major-hurricane “drought” became so widespread that researchers took pains to show how natural variability could lead to periods of a decade or longer without major landfalls.

Sure enough, the tide turned when fast-strengthening Harvey struck the Texas coast in August 2017, leading to 89 deaths and estimated $133 billion in damage, the most for any U.S. hurricane after inflation except for Katrina. Harvey has since been followed by major U.S. landfalls from Irma (Category 3, southwest Florida, 2017), Michael (Category 5, Florida Panhandle, 2018), Laura (Category 4, southwest Louisiana, 2020), and Zeta (Category 3, southeast Louisiana, 2020).

With any luck, the inevitable next EF5 tornado to hit the United States — whenever that may be — won’t leave this kind of awful legacy.

Also see:  Confronting the ‘isolated, lonely, painful deaths’ from heat waves

Website visitors can comment on “Eye on the Storm” posts. Please read our Comments Policy prior to posting. Comments are generally open for 30 days from date posted. Twitter: @DrJeffMasters and @bhensonweather