No EF6 rating exists on the Enhanced Fujita Scale, and no meteorological organization has plans to create one. The scale tops out at EF5, which corresponds to estimated wind speeds of 200 mph and above and describes damage so catastrophic that well-built frame houses are swept clean off their foundations. The reason for the ceiling is more interesting than it first sounds: the EF scale is fundamentally a damage scale, not a wind-speed scale, and once a building is completely destroyed, there is no way to tell from the wreckage whether the winds that leveled it were 210 mph or 310 mph. The rubble looks the same.
A Damage Scale Dressed Up as a Wind Scale
The original Fujita Scale, introduced by Ted Fujita in 1971, assigned tornado ratings from F0 through F5 based on the type and severity of damage observed. It was revised in 2007 into the Enhanced Fujita Scale, which refined the damage indicators and tightened the estimated wind-speed ranges tied to each rating. But the core method stayed the same: survey teams from the National Weather Service go out after a tornado, look at what was destroyed and how, and match the destruction to a catalog of damage indicators and degrees of damage. Those observations are then converted into an estimated wind speed, which determines the EF rating.
This means the rating is always indirect. Nobody is standing in the path of an EF5 tornado with an anemometer. The wind speeds you see listed for each EF category are estimates derived from engineering analysis of how much force it takes to produce specific kinds of structural failure. EF5 begins at an estimated 200 mph, but the scale has no upper bound for wind speed within that category. Whether the actual winds were 205 or 295 mph, if the damage looks like total destruction of a sturdy structure, the tornado gets an EF5.
An EF6 category would require surveyors to identify a damage state worse than “completely destroyed.” But there is no meaningful structural distinction beyond that point. A house reduced to a bare slab with debris scattered hundreds of yards away cannot be more destroyed. The damage indicators simply run out of room. This is the practical reason the scale has a ceiling, and it has nothing to do with atmospheric physics imposing a speed limit on tornadoes.
Radar Shows Winds That Blow Past 200 mph
While the EF scale stops at 200 mph, actual tornado winds do not. Mobile Doppler radar units, particularly the Doppler on Wheels (DOW) fleet operated by research teams, have recorded tornado wind speeds that far exceed the EF5 threshold. The fastest radar-measured tornado winds reach roughly 140 meters per second, which is about 313 mph.1Communications Earth & Environment. The strongest winds in tornadoes are very near the ground A separate analysis of DOW-observed tornadoes found a maximum wind speed of 144 meters per second, or about 322 mph, measured at just 37 meters above the ground.2PubMed Central. Supercell tornadoes are much stronger and wider than damage-based ratings indicate
Those are staggering numbers. For context, the EF5 threshold of 200 mph is already enough to hurl cars through the air and strip asphalt off roads. Winds exceeding 300 mph represent a regime of destruction that the damage scale was never designed to differentiate. And yet these measurements come with an important caveat: radar beams typically sample the air well above the surface, often more than 50 meters up, which is considerably higher than the rooftops of most buildings.1Communications Earth & Environment. The strongest winds in tornadoes are very near the ground Wind speeds at the actual ground level, where structures sit, can differ from what radar measures aloft due to friction and turbulence near the surface. Still, research increasingly suggests the strongest winds may actually be concentrated very close to the ground, which makes the gap between radar readings and the damage-based scale even more striking.
The El Reno Tornado and the Rating Disconnect
If you want a single event that crystallizes the tension between radar-measured winds and damage-based ratings, the May 31, 2013 El Reno, Oklahoma tornado is the case meteorologists keep returning to. At its widest, the tornado stretched about 2.6 miles across, making it the widest tornado ever recorded in the United States. Mobile radar units captured extraordinary wind data from the storm. A dual-Doppler wind analysis near the location where several storm chasers were killed resolved ground-relative wind speeds exceeding 90 meters per second, which is above the minimum threshold for EF5 damage.3Monthly Weather Review. Aerial Damage Survey of the 2013 El Reno Tornado Combined with Mobile Radar Data
Yet the El Reno tornado was officially rated EF3. Why? Because it tracked largely through open farmland. There were very few substantial structures in its path for survey teams to assess, so the highest damage indicators were never triggered. The tornado was almost certainly stronger than EF3, and the radar data confirmed winds consistent with EF5 intensity in at least part of the vortex, but the rating system could only go as high as the damage evidence allowed. This is not an obscure academic complaint. It means some of the most powerful tornadoes in history may carry ratings that dramatically understate their true intensity, simply because they happened to miss populated areas or sturdy buildings.
How Often Tornadoes Get Underrated
El Reno is the most famous example, but the underrating problem is systematic. The Enhanced Fujita scale relies on damage as a proxy for wind speed, and that proxy breaks down whenever the tornado’s path does not contain the right structures. A tornado that tears across open prairie, a forest, or a neighborhood of mobile homes may produce catastrophic damage that looks like one rating while the winds were consistent with something much higher. Research into this mismatch has used analytical wind-field models paired with engineering fragility curves to simulate what damage a tornado of a known wind speed would produce, and then compare that to how the damage would be interpreted by survey teams. The findings confirm what storm researchers have long suspected: the damage-to-wind-speed conversion is not reliable enough to capture true tornado intensity, particularly for rural tornadoes or those that strike areas without the damage indicators the scale was built around.4American Meteorological Society. An Approach for Assessing Misclassification of Tornado Characteristics Using Damage
This matters beyond the rating itself. Tornado climatology, which tracks how tornado characteristics change over time and geography, depends on damage ratings. If rural and open-terrain tornadoes are systematically underrated, then our picture of tornado strength distribution is skewed. Regions with fewer structures per square mile may appear to have weaker tornadoes than they actually do, and trends over time can be distorted if development patterns change what structures are available for damage assessment.
Research using DOW data has gone further, finding that supercell tornadoes in general tend to be considerably stronger and wider than their damage-based ratings suggest.2PubMed Central. Supercell tornadoes are much stronger and wider than damage-based ratings indicate The implication is that the EF scale does not just cap out at EF5 because of a missing category. It routinely understates tornado intensity across the board because the damage proxy is inherently lossy.
The International Fujita Scale
The limitations of the EF scale have prompted work on alternatives, particularly outside the United States. The European Severe Storms Laboratory has developed the International Fujita (IF) scale, which expands the number of applicable damage indicators well beyond those used in the American EF system. The IF scale accounts for building sturdiness variations across different countries, tree characteristics, and commonly affected objects like vehicles and outdoor furniture.5Natural Hazards. Revisiting a 25-year database of tornado damage paths using the new International Fujita (IF) scale
This is a meaningful improvement because the EF scale’s damage indicators were calibrated primarily for American construction standards and building types. A well-built frame house in Oklahoma and a brick row house in northern France fail in different ways under the same wind load, and the EF scale’s indicators do not account for that. The IF scale’s broader set of indicators makes it more applicable in regions where the building stock, vegetation, and landscape differ from the American Midwest.
However, the IF scale still does not add a sixth category. It retains the same 0-to-5 range because the fundamental ceiling problem remains: once everything at a site is completely destroyed, there is no damage-based way to distinguish higher wind speeds. The IF scale is better at rating tornadoes accurately within the existing categories, especially in international contexts, but it does not extend the top of the scale.
Would Adding EF6 Require a Different Approach Entirely?
If the meteorological community ever decided an EF6 category was worth creating, it could not rely on damage alone. The rating would need to incorporate direct wind measurements, most likely from mobile radar, as a primary input rather than just a research curiosity that sits alongside the official damage survey. Some researchers have argued for exactly this kind of hybrid approach: use damage indicators where available, but allow radar-measured wind speeds to override or supplement the damage rating when instrumented data exist.
The practical obstacles are significant. Mobile Doppler radar units are expensive to deploy and can only be in one place at a time. Most tornadoes are never observed by a DOW or similar platform. The coverage is deeply uneven: well-studied tornado events in Oklahoma and Kansas may have excellent radar data, while equally powerful tornadoes in the Southeast or at night go unobserved. Creating a rating category that depends on instrumented observations would mean most tornadoes could never be evaluated for that category, which defeats the purpose of a universal classification system.
There is also a calibration problem. Radar measures wind speeds at a specific height above the ground, not at the surface where damage occurs. The relationship between winds aloft and winds at ground level is complex and varies with terrain, surface roughness, and the tornado’s own structure. Translating a radar-measured 300 mph wind at 40 meters up into a reliable ground-level estimate is still an active area of research, and the uncertainty is large enough that pinning a new damage category to it would be premature.
Why People Keep Asking About EF6
The question tends to spike in public interest after major tornado events, especially when media coverage describes a tornado as “off the charts” or emphasizes record-breaking wind speeds. There is a natural human impulse to think that if something is the worst category, there must be something even worse lurking beyond it. Hurricane categories work the same way in public imagination: after every devastating Category 5 hurricane, calls for a Category 6 emerge.
But the analogy is imperfect. The Saffir-Simpson hurricane scale is based on sustained wind speed measurements, not damage proxies, so adding a Category 6 would at least be conceptually straightforward even if meteorologists have resisted it. The EF scale’s dependence on damage means an EF6 is not just an extension of the same method. It would require a fundamentally different kind of evidence, which is why the conversation among researchers is less about “should EF6 exist” and more about “should the rating system incorporate direct wind measurements at all.”
The confusion is compounded by how EF ratings are communicated. When the news says “EF5 tornado with winds over 200 mph,” many people reasonably interpret EF5 as meaning approximately 200 mph. The reality is that EF5 means “at least 200 mph, possibly much more, and we cannot tell from the damage how much more.” A tornado rated EF5 could have peak winds of 210 mph or 310 mph, and the rating would be the same. The scale is not a speedometer with precise readings; it is a damage classification that correlates loosely with wind speed.
Where the Strongest Winds Actually Live Inside a Tornado
One of the more surprising findings from recent radar research is that the most extreme winds in a tornado may be concentrated in a thin layer very close to the ground. Conventional radar measurements, even from mobile units parked near a tornado, typically sample the vortex at heights above 50 meters, well above the roofs of most buildings.1Communications Earth & Environment. The strongest winds in tornadoes are very near the ground If the peak winds are actually in the lowest few tens of meters, then even the impressive radar measurements of 300+ mph may be underestimates of what is happening at the surface.
This matters for understanding why damage sometimes looks “worse than the rating” even in well-surveyed urban tornadoes. If the most destructive winds hug the ground in a shallow layer that radar cannot fully resolve, the damage pattern might reflect winds higher than anything the radar captured. It also means the engineering models used to link wind speed to structural failure might need recalibration, because they were developed using wind speed estimates that may themselves be too low.
The research is still evolving. Newer radar technologies with finer vertical resolution and lower scan angles are beginning to fill in this gap, but the dataset of tornadoes observed with these instruments is still small. Every well-instrumented tornado that researchers capture adds to the picture, and the trend so far suggests that the gap between what the EF scale rates and what actually happens at the surface is larger than previously thought.2PubMed Central. Supercell tornadoes are much stronger and wider than damage-based ratings indicate
What the Scale Is Actually Good For
For all its limitations, the EF scale serves a purpose that no radar-based system currently can: it provides a standardized, after-the-fact rating for every tornado that hits an area with structures, anywhere in the country, using survey teams and photographic evidence rather than expensive mobile instruments. It gives emergency managers, insurance adjusters, and climatologists a common language. When a forecast office issues a damage survey with an EF rating, responders know roughly what level of destruction to expect and how to allocate resources.
The scale also does a reasonable job at the lower end. The difference between EF0 and EF2 damage is visually distinct and practically meaningful for response planning. The problems become most acute at the upper end, where the jump from EF3 to EF5 can hinge on whether the tornado happened to hit a well-built home or a strip of farmland. And at the very top, where EF5 absorbs everything from 200 mph to whatever the atmosphere can produce, the rating becomes a floor rather than a measurement.
For everyday purposes, knowing whether a tornado was EF3 or EF4 matters for building codes, insurance claims, and community planning. Knowing whether a tornado’s peak winds were 250 or 320 mph is, at the moment, mostly of interest to researchers studying vortex dynamics. If that ever changes, if engineering or warning systems need that granularity to save lives, the conversation about extending or replacing the scale will shift from academic debate to practical necessity. For now, the scale stops at EF5 because the damage stops telling us anything new, even when the winds clearly do not.