Twister Aside, How Do They Actually Measure the Strength of Tornadoes?

On the afternoon of March 18, 1925, residents of southeastern Missouri watched with dread as a line of pitch-black storm clouds gathered on the horizon. As torrential rain and hail began to fall, a cluster of 12 sinister funnel clouds snaked their way down to the ground and began moving northeast, into neighbouring states of Illinois and Indiana. One of these tornadoes, travelling at 117 kilometres an hour, traversed an unprecedented 352 kilometres, sweeping through 14 counties and 19 major communities, destroying more than 15,000 homes and inflicting some $17 million in damage – almost $300 million in today’s money. By the time the skies cleared three hours later, 747 people lay dead and more than 13,000 injured. Nearly a third of the victims were children, killed when nine schools collapsed in the 300 kilometre-per-hour winds. To this day, the 1925 Tri-State Tornado Outbreak remains the single deadliest and most destructive tornado event in United States history, and the third most extreme worldwide.

If you have seen the 1996 disaster film Twister, then you know that tornadoes are rated according to F-numbers, with F5s – like the 1925 Tri-State Tornado – being the most powerful. But what do these numbers actually mean, and how does one measure and classify something as destructive and chaotic as a tornado? Well, dive into the nearest cellar and watch out for flying cows as we delve into the science of one of nature’s most violent weather events, and the surprisingly fascinating history of measuring them, along with why exactly they chose “F” for the scale.

To begin with, tornadoes form under particular meteorological conditions, wherein a combination of warm, moist air on the ground and cooler, drier air within a storm cloud promotes the formation of strong updrafts. The presence of wind shear – a sudden change in wind direction and speed at altitude – promotes the formation of a rotating vortex, shaping the updraft into the iconic funnel cloud. While nearly all thunderstorms have the potential to produce tornadoes, the strongest are formed by so-called supercells, severe, long-lived storms which contain a powerful vortex or mesocyclone that can grow upwards through the storm and downwards to the ground, often spawning multiple highly-destructive funnels. And while tornadoes occur on every continent, the geography and climate of the American Midwest – AKA “tornado alley” – make this region particularly susceptible to this phenomenon, with an average of around 150 tornadoes being recorded every year. Specifically, the convergence of cool, dry air from the Rocky Mountains; warm, moist air from the Gulf of Mexico; and warm, dry air from Texas – combined with the region’s mostly flat terrain – create the ideal conditions for tornado formation.

Yet while humans have had to deal with tornadoes since the dawn of history, it was not until the 1950s that meteorologists truly began to study and understand this phenomenon – and not until the 1970s that they developed a reliable method for measuring their strength. The man responsible for the tornado classification scale still used to this day was pioneering Japanese-American researcher Tetsuya “Ted” Fujita – AKA “Mr. Tornado.”

Ted Fujita was born on October 23, 1920 in the village of Sone on the Japanese island of Kyushu. As a child, Fujita had a passion for geology and geography, exploring caves and volcanoes in his spare time and drawing detailed topographical maps of his surroundings. Upon graduating from high school, however, he decided to pursue a career in engineering and planned to enroll at Hiroshima University. In a twist of fate, however, Fujita’s father convinced him to attend Meiji College instead. This most likely saved his life, for Hiroshima University lay close to ground zero of the first atomic bomb dropped on August 6, 1945. After graduating from Meiji College in 1943 with a degree in Mechanical Engineering, Fujita landed a teaching position in the city of Kokura. Here, on August 9, 1945, he almost had a brush with an atomic bomb as air raid sirens blared across the city, warning of an approaching flight of American B-29 bombers. However, the sky over Kokura that day was overcast, the visibility made even worse by thick smoke from the nearby city of Yahata, which had been badly firebombed the day before. So, after circling Kokura for nearly an hour, the bombers abandoned their primary target and instead dropped the second atomic bomb on the city of Nagasaki, 216 kilometres away. Six days later, the Japanese government announced its unconditional surrender. The Second World War was over.

A month after the end of the war, Fujita and a group of students were sent to Nagasaki to survey the damage inflicted by the atomic bomb. By measuring the starburst patterns of felled trees and lampposts created by the explosion’s powerful downdrafts, Fujita was able to determine that the bomb had been detonated at an altitude of 160 metres – the optimum height for maximum blast damage. Though he didn’t know it at the time, these observations would later prove crucial to one of his most important discoveries.

In 1946, Fujita decided to turn his focus to meteorology, and applied for a Department of Education grant to instruct other teachers about the weather. In his spare time, he conducted his own meteorological research by climbing mountains to observe weather patterns from above. In this manner he made a number of key insights into the dynamics of thunderstorms – especially the formation of cold air downdrafts. However, he soon discovered that the American Thunderstorm Project – a joint research effort between the United States Weather Bureau and National Advisory Committee – had already beaten him to the punch. But Fujita persevered, and in September 1948 he had his first encounter with the phenomenon that would make his career. After a tornado made landfall at Enoura on Kyushu Island, Fujita rushed to the scene and walked along the entire path of the storm, making detailed notes of the damage patterns. Realizing that his observations might be worth publishing, Fujita bought a typewriter – which cost two and a half times his monthly salary – and began painstakingly translating his work into English for western scientific journals. At around this time, Fujita met meteorologist Dr. Horace Byers at the University of Chicago, with whom he corresponded while working on his PhD thesis at the University of Tokyo. Shortly after receiving his doctorate in 1953, Fujita accepted an offer from Byers to come to Chicago, where he would spend the rest of his career. It was a timely transition, for the ever-worsening inflation and food shortages in postwar Japan had begun to weight heavily on Fujita, pushing him into an ever-deeper depression – and for more on an unexpectedly successful product of postwar Japanese scarcity, please check out our video The Surprisingly Interesting Origin of Instant Ramen on another of our sister channels, Origins.

Once in Chicago, Fujita began focusing almost exclusively on the study of tornadoes. Ironically, though he would eventually come to be known as “Mr. Tornado”, Fujita only actually saw three tornadoes in his lifetime – all on June 12, 1982 in Colorado. Instead, Fujita pioneered methods for studying and categorizing tornadoes by examining the trail of destruction they leave behind. He became especially skilled at the practice of photogrammetry – using aerial photographs of a tornado-affected area to construct weather maps and calculate different parameters like wind speed. Such indirect observations are unfortunately a fact of life for tornado researchers, since the odds of a tornado passing close to an established weather station are very low. Also, most tornadoes are so short-lived that by the time storm chasers get to the scene and set up their instruments, the tornado has likely already dissipated. Nonetheless, Fujita used these methods to great effect on several major weather events including the 1957 Fargo, North Dakota tornado; the 1965 Palm Sunday tornado outbreak in Iowa, Ohio, Michigan, and Indiana; and the 1970 Lubbock, Texas tornado – obtaining surprisingly accurate data. This is despite the fact that throughout his career, Fujita refused to use computers, believing them incapable of truly understanding the complex dynamics of weather systems.

Based on his research in the 1950s and 60s, in 1971 Fujita introduced a standardized scale for rating tornado intensity, known as the Fujita scale or F-scale. Originally, Fujita developed a 13-point system to connect the Mach Number scale – which measures fluid velocities against the local speed of sound – with the Beaufort scale. Developed in 1805 by Irish hydrographer Francis Beaufort, the Beaufort scale measures wind speed according to its effect on everyday objects. For example, 0 on the scale indicates no wind – water is glassy and smoke rises vertically; 6 indicates a strong breeze – whitecaps form on waves and large tree branches begin swaying; 9 indicates gale-force winds – foam forms on waves and roof tiles and chimney pots on houses are damaged; and 12 indicated hurricane-force winds, capable of flattening houses and other structures. In Fujita’s original scale, F0 corresponded with Level 8 on the Beaufort scale – that is, 104-136 kilometre per hour winds – F1 with Level 12 or 138-176 kilometres per hour; and F12 with Mach 1. But as no tornado was known to have exceeded 500 kilometres per hour, in practice only levels F0-F5 were meant to be used – though Fujita did set aside an extra F6 designation for what he called “inconceivable tornadoes” more powerful than F5s. Yet while several storms were initially given F6 designations – including the 1970 Lubbock; 1995 Pampa, Texas; and the 1974 Xenia, Ohio tornadoes – these have since been officially downgraded to F5. Nonetheless, F5 tornadoes remain exceedingly rare; of the 150 or so tornadoes recorded in the United States every year, 76% are F0s or F1s, 20% are F2s or F3s, 3% are of undetermined intensity, and only 1% are F4s or F5s.

Each level of the Fujita scale is based on the estimated wind speed of a 3-second gust in a particular area, calculated from the type of damage inflicted in that area. For example, an F0 tornado, with wind speeds of of 104-136 kilometres per hour, will cause only light damage, ripping shingles off roofs, uprooting small trees, and damaging light structures like garden sheds. An F2 tornado, by contrast, with wind speeds of 180-250 kilometres per hour, can inflict significant damage, ripping roofs off houses, overturning cars, and uprooting medium trees; wile an F5 tornado, with wind speeds of 419-512 kilometres per hour, can destroy all but reinforced concrete buildings, throw cars thousands of metres, and uproot large trees and even grass.

Two years later in 1973, Allen Pearson of the National Severe Storms Forecast Center enhanced Fujita’s scale by adding additional parameters to indicate a tornado’s path length and width, with each number on the 0-5 scale indicating a certain range of distances. In this system, a tornado is given one Fujita rating and two Pearson ratings, such that, for example, an F4 category tornado with a path length of 100 kilometres and a path width of 700 metres would be rated F,P,P 4,4,4. However, the Pearson scale never really caught on, and today the path length and width are simply noted directly next to the Fujita category.

Once the Fujita scale was officially adopted, the National Oceanic and Atmospheric Administration or NOAA used it to retroactively rate U.S. tornadoes going back to 1950. And in the 1970s, the Nuclear Regulatory Commission carried out a study to gauge tornado damage risk to nuclear power plants, a project which ultimately yielded a list of tornadoes rated F2 or higher going back to the year 1871 and all tornadoes resulting in deaths going back to 1680.

But while the Fujita scale was the first of its kind and used all over the world for decades, it was eventually found to have significant shortcomings. For example, the number of damage indicators used to make wind speed calculations was too low, while engineering studies on the effects of wind on various structures revealed that the original Fujita scale wind speeds did not accurately match real-world damage. So, on February 1, 2007, the original Fujita scale was retired and replaced with the Enhanced Fujita or EF Scale. In practical terms the EF-scale is nearly identical to the older F-scale, using the same F0-F5 ratings and their corresponding damage levels. However, the wind speeds associated with these damage levels have been updated, based on an expanded list of 28 damage indicators. These indicators describe the types of structures likely to be damaged or destroyed at a given wind speed – such as farm outbuildings, motels, big box stores, 5-20-storey high rises, and free-standing electrical transmission line towers.

Despite these enhancements, however, the Fujita scale remains highly subjective, with the accuracy of a given rating depending on the skill and experience of the surveyor. According to The Tornado Project, a nonprofit tornado research organization:

“…the less experienced the surveyor is, the more likely he/she is to be awed by the damage, and the more likely they are to give it a high rating…Media hype and inexperience with tornado damage also plays a big part in exaggerated F-Scale claims seen on television or in the paper. A reporter may see a collapsed concrete block home and be very impressed, never noticing that there was no mortar between the blocks. They may be aghast to see a park whose trees have been levelled, but not know that the species had very shallow roots, planted in soil that was soft and soggy from torrential rains, and thus easily toppled. They may see a roof that had been blown a quarter of a mile from its house, and not know that the roof was attached to the house with only a few nails, and when lofted into the air, acted as a “sail.” They may see a light post that is bent at a 30 degree angle and think that it must have taken a 600 mph wind to do that, not knowing that a van had been blown into the pole, bending it, then been towed off to help clear the streets.”

Indeed, since tornadoes can only be rated based on the actual damage they inflict, tornadoes that would be given one rating in a certain location might be given another rating in a different location. For example, the tornado that touched down near Seymour, Texas on April 10, 1979 would likely have been rated at F4 had it actually passed through the town. However, as it only inflicted damage on trees and telephone poles, it could only be rated at F2. It is also worth noting that the F-scale rating of a tornado does not necessarily correlate with its physical size. Large tornadoes can be weak and small tornadoes strong; tornadoes can also increase and decrease in size throughout their life cycle. Single large tornadoes can even contain multiple subvoritices within them, which can result in strange phenomena like a tornado destroying one house while leaving the neighbours untouched. Interestingly, when Fujita first proposed the existence of subvortices in the mid-1970s, meteorologists dismissed his ideas. He was later proven to be absolutely correct.

Though Ted Fujita’s name has become forever associated with tornadoes, he made other major contributions to meteorology. In the mid-1950s, for example, he pioneered the practice of plotting sharp pressure jumps on barometer readings to predict large tornado outbreaks – a technique known as mesoanalysis. And in 1975, he made a major discovery that helped solve an aviation mystery. On June 24 of that year, Eastern Airlines Flight 66, a Boeing 727 flying from New Orleans to New York, crashed while landing at John F. Kennedy International Airport. Near the end of its approach, the aircraft experienced a severe downdraft, causing it to slam into the ground just short of the runway and burst into flames – killing 113 of the 124 people aboard. The accident baffled investigators, for while there were severe thunderstorms in the area, no other aircraft at the airport that day encountered the same downdrafts. But the investigation gained a new lead when Homer Mouden, a safety expert with the Flight Safety Foundation, asked Ted Fujita for his advice. At the time, Fujita was studying a strange phenomenon he had observed in the wake of the April 1974 “Jumbo” outbreak in the northeastern United States: radial “starburst” pattens of toppled trees. These starbursts reminded him of similar patterns he had seen below the hypocentre of the Nagasaki atomic bomb blast, and led him to hypothesize the existence of small but powerful downdrafts of rain-cooled air that suddenly plunge down from thunderclouds. Fujita named these phenomena microbursts, and after studying airport weather data and flight 66’s flight data recorder, concluded that the crash had been caused by one of these downdrafts. This discovery led to the implementation a host of new safety measures such as special training for pilots and the installation of doppler weather radars at airports – measures which have likely saved hundreds if not thousands of lives.

After an illustrious career spanning nearly five decades, Ted Fujita finally retired from the University of Chicago in 1990, dying eight years later at the age of 78. Though known around the world as “Mr. Tornado”, his name forever immortalized by his eponymous tornado-rating scale, Fujita remained modest and humble, once remarking:

“…even if I am wrong 50% of time, that would still be a tremendous contribution to meteorology.”

Expand for References