The El Reno tornado of May 31, 2013, killed eight people and injured more than a hundred, yet its physical destruction was strikingly modest for what turned out to be the widest tornado ever recorded in the United States. At 2.6 miles across and packing radar-measured winds well above 200 mph, the tornado’s path ran mostly through open farmland west of Oklahoma City, sparing the dense suburban development that lay just to its east. The gap between what this tornado was capable of and what it actually destroyed remains one of the most studied paradoxes in modern severe-weather science.
Why Physical Damage Was Limited Despite Extreme Winds
The El Reno tornado touched down around 6:03 p.m. CDT southwest of El Reno, Oklahoma, and tracked generally east-northeast for roughly 40 minutes before lifting. Most of that path crossed agricultural land, pastures, and sparsely populated rural stretches of Canadian County. Farmsteads, outbuildings, and scattered homes took damage along the track, but the tornado never made a sustained pass through a built-up neighborhood or commercial district. That mismatch between ferocity and destruction is central to understanding why the event produced relatively little insured loss compared to its meteorological significance.
For context, the Moore tornado had struck just eleven days earlier on May 20, 2013, killing 24 people and inflicting over two billion dollars in damage as it plowed through a densely populated suburb. The El Reno tornado was, by most objective measures, a far more powerful storm. But because it churned across open ground, total property losses stayed in the tens of millions rather than billions. Tornado damage, as a general rule, is determined less by a storm’s peak wind speed and more by what stands in its way.
Fatalities and the Storm Chaser Tragedy
Eight people died in or because of the El Reno tornado, including three experienced storm researchers. Tim Samaras, a veteran tornado scientist known for deploying in-situ measurement probes, was killed along with his son Paul Samaras and colleague Carl Young when the tornado’s rapid changes in direction and multiple-vortex structure caught them in its path. Their deaths shocked the severe-weather community and prompted serious reflection about the risks storm chasers accept in the field.
The other fatalities included motorists who were overtaken by the tornado on area roads. A dual-Doppler wind analysis near the locations where several storm chasers died resolved ground-relative wind speeds exceeding 90 meters per second, which is above the minimum threshold associated with EF5 damage. In other words, the winds that struck those vehicles were among the most violent a tornado can produce.1Monthly Weather Review. Aerial Damage Survey of the 2013 El Reno Tornado Combined with Mobile Radar Data More than 150 other people sustained injuries, many of them motorists caught on highways as the tornado shifted course unpredictably.
The EF3 Rating and Why It Remains Controversial
After ground and aerial surveys, the National Weather Service rated the El Reno tornado EF3 on the Enhanced Fujita scale, corresponding to estimated winds of 136 to 165 mph. That rating has been a source of frustration among researchers ever since. The EF scale rates a tornado based on the worst damage it causes to specific types of structures, called damage indicators. When a tornado spends most of its life over open fields with few or no buildings to destroy, surveyors have little to work with. You cannot rate what a tornado did to homes if it never hit any homes.
Mobile radar instruments told a dramatically different story. The rapid-scan X-band polarimetric radar known as RaXPol recorded radar-relative radial velocities exceeding 135 meters per second, roughly 302 mph, in rural areas that were essentially devoid of structures from which damage ratings could be made.2Weather and Forecasting. Some Considerations for the Use of High-Resolution Mobile Radar Data in Tornado Intensity Determination That figure places the winds deep into what the original Fujita scale would have classified as F5 territory. The disconnect between the radar evidence and the official EF3 rating is not a quirk of this one storm. Research has shown that damage-estimated tornado wind speeds tend to be systematically lower than radar-estimated near-surface wind speeds, and the gap grows wider for stronger tornadoes. One analysis found this underestimation exists across the full spectrum of land-use density, not just in the most rural areas, and suggested the legacy F-scale wind speed ranges may ultimately be more accurate for rating strong and violent tornadoes than the current EF-scale thresholds.3Monthly Weather Review. Comparison of Tornado Damage Characteristics to Low-Altitude WSR-88D Radar Observations and Implications for Tornado Intensity Estimation
The practical consequence of this rating gap is significant. An EF3 tornado gets logged in the historical record as a “severe” event, while an EF5 is classified as the most extreme category of tornado. For El Reno, the EF3 label understates what the atmosphere was actually doing. Researchers who study the storm routinely describe it as a violent tornado capable of EF5-level destruction, and its official rating stands more as a limitation of the rating system than a reflection of the tornado’s true intensity.
A Multiple-Vortex Monster
Part of what made the El Reno tornado so dangerous, and so difficult for storm chasers to navigate, was its complex internal structure. It was not a single funnel spinning in one spot. Rapid-scan radar data collected every two seconds revealed multiple secondary vortices orbiting within the larger circulation, each one a smaller but intense tornado-within-a-tornado.4Monthly Weather Review. The Multiple-Vortex Structure of the El Reno, Oklahoma, Tornado on 31 May 2013
These secondary vortices typically formed inside the radius of maximum wind, mostly in the left-rear part of the circulation relative to the tornado’s direction of travel. They would then sweep around the center and dissipate in the forward quadrants. Some persisted for at least a full minute, long enough to carve out concentrated streaks of extreme damage within the broader swath.4Monthly Weather Review. The Multiple-Vortex Structure of the El Reno, Oklahoma, Tornado on 31 May 2013 The subvortices moved fast and erratically relative to the parent tornado’s motion, which meant that a person or vehicle positioned just outside what appeared to be the tornado’s edge could suddenly be engulfed by a sub-vortex swinging outward.
This structure had direct consequences for both damage and casualties. The aerial damage survey combined with mobile radar data revealed a tornadic debris signature roughly two kilometers wide during the analysis period, an enormous footprint reflecting how much material the storm was lofting.5Monthly Weather Review. Photogrammetric Analysis of the 2013 El Reno Tornado Combined with Mobile X-Band Polarimetric Radar Data Even in areas where the main vortex caused moderate surface-level damage, individual sub-vortices left narrow corridors of far more intense destruction, scattering debris across a wider area than the main path would suggest.
The Evacuation Crisis
The El Reno tornado nearly produced a catastrophe far worse than what actually occurred. As the storm bore down on the western edge of the Oklahoma City metropolitan area, several local television broadcasters broke from standard shelter-in-place messaging and urged residents to get in their cars and drive south to escape the tornado’s path. This advice was well-intentioned but triggered a massive, uncoordinated evacuation. Interstates and major highways south of Oklahoma City quickly became gridlocked with tens of thousands of vehicles.6Weather, Climate, and Society. A Spatiotemporal Perspective on the 31 May 2013 Tornado Evacuation in the Oklahoma City Metropolitan Area
Had the tornado maintained its eastward track and reached those clogged highways, the death toll could have climbed into the hundreds. Thousands of people sat motionless in bumper-to-bumper traffic with no ability to maneuver, no below-ground shelter within reach, and a 2.6-mile-wide tornado approaching from the west. The tornado weakened and dissipated before it reached the worst of the congestion, but the near-miss reshaped how emergency managers and broadcast meteorologists in tornado-prone regions think about evacuation advice. The consensus that emerged from the aftermath was clear: telling a metro population to flee a tornado by car is more likely to kill people than save them, because the resulting traffic jam traps far more people than would have been at risk sheltering in place.
The evacuation episode is arguably the most consequential piece of “damage” the El Reno tornado inflicted, even though no one died in the traffic jams. It exposed a vulnerability in tornado response that had not been tested at that scale before and forced an uncomfortable reckoning with how public communication during severe weather events can itself become a hazard.
What Would Have Happened in a City
One of the reasons scientists have spent so much time analyzing the El Reno tornado is that it offers a window into what a worst-case urban tornado strike could look like. A 2.6-mile-wide tornado with EF5-capable winds hitting a dense metropolitan area would produce damage on a scale that no American city has experienced in modern history. The Moore tornado eleven days earlier destroyed thousands of homes and caused massive casualties across a path that was about a mile wide and 17 miles long. The El Reno tornado was more than twice as wide, with higher measured wind speeds, and it was heading toward a far more populated area before it lifted.
The storm’s actual damage footprint, while locally devastating for rural residents who lost homes, livestock, and livelihoods, remained a fraction of its potential. The structures it did hit were mostly agricultural buildings, vehicles, fences, and a limited number of residences. That small sample of damaged structures was all the NWS surveyors had available to estimate wind speeds from, which circles back to why the official rating feels so inadequate to researchers who watched the radar data in real time.
Damage Beyond Buildings
Tornadoes inflict damage that extends well beyond what property-loss figures capture. The psychological toll on survivors, first responders, and even the broader community of storm chasers and meteorologists was substantial in the wake of El Reno. The deaths of the Samaras team in particular reverberated through the storm-chasing community and the broader field of severe-weather research, prompting new conversations about field safety protocols and the ethics of tornado intercept operations.
Research on tornado-affected populations more broadly has documented a range of mental health consequences including PTSD, anxiety, depression, and substance abuse in both children and adults. One systematic review found that among roughly 2,000 adolescents who experienced major tornadoes, about five percent reported suicidal thoughts, and the majority of tornado survivors had not accessed any mental health support even two and a half years after the event.7PubMed Central. Mental Health Impacts of Tornadoes: A Systematic Review Those findings come from studies of other tornado events, but they illustrate a pattern that applies to any community struck by a violent tornado: the visible wreckage gets cleaned up in weeks or months, while the psychological wreckage can persist for years with little formal support.
For El Reno specifically, the psychological dimension extended to people who were never in the tornado’s path at all. The hundreds of thousands of Oklahoma City residents who spent that evening trapped in traffic, watching the storm approach on their phones and car radios, experienced genuine terror. Many had survived the Moore tornado less than two weeks earlier. The cumulative stress of living through back-to-back historic tornado events during the spring of 2013 left a mark on the region’s psyche that is harder to quantify than property losses but no less real.
How El Reno Changed Tornado Science and Policy
The El Reno tornado became one of the most intensively studied tornadoes in history, and its legacy has shaped both the science of tornado measurement and the practice of public warning. On the measurement side, the storm laid bare the limitations of rating tornadoes purely by what they do to buildings. When a tornado’s strongest winds never encounter a well-built structure, the EF scale essentially cannot see them. The RaXPol radar data from El Reno, showing 135-meter-per-second winds over empty farmland, became a landmark example in the ongoing debate about whether radar data should supplement or even replace damage surveys for intensity ratings.2Weather and Forecasting. Some Considerations for the Use of High-Resolution Mobile Radar Data in Tornado Intensity Determination
The photogrammetric analysis of the tornado, combining visual imagery with radar observations, also advanced the field. Researchers were able to map the relationship between the visible funnel, the radar-detected debris cloud, and the internal wind field in ways that had not been possible before, partly because the tornado was so large and partly because so many instruments were pointed at it simultaneously.5Monthly Weather Review. Photogrammetric Analysis of the 2013 El Reno Tornado Combined with Mobile X-Band Polarimetric Radar Data The two-kilometer-wide debris signature detected in polarimetric radar data gave researchers new benchmarks for understanding how large tornadoes loft and distribute debris, which in turn feeds into better real-time tornado detection.
On the policy side, the evacuation debacle forced a rethinking of broadcast messaging during tornado warnings. The National Weather Service and the broadcast meteorology community spent years after 2013 refining guidance to discourage mass vehicular evacuation during tornado emergencies in metropolitan areas. The lesson was expensive in terms of public trust and could easily have been expensive in lives: shelter in a sturdy structure is almost always safer than fleeing in a car, especially when hundreds of thousands of other people are fleeing at the same time.
Counting Damage That Does Not Fit on a Spreadsheet
If you measure the El Reno tornado’s damage strictly in dollars and destroyed structures, it looks like a moderate event. Tens of millions in property losses, a handful of destroyed or heavily damaged rural homes, agricultural infrastructure torn apart. By those metrics, it ranks well below the Moore tornado, the Joplin tornado of 2011, or many other famous severe-weather disasters. But that framing misses what made the event so consequential.
Eight people died, including three of the most respected figures in tornado field research. An entire metropolitan area narrowly avoided a mass-casualty highway disaster. The scientific community gained definitive evidence that the EF scale can dramatically undercount a tornado’s true intensity, a finding that has implications for how tornado risk is communicated and how building codes are calibrated. And a region already reeling from the Moore tornado absorbed another round of trauma that compounded existing stress.
The El Reno tornado was, in some respects, the most informative tornado disaster in American history precisely because the worst-case scenario did not fully materialize. It showed what was possible without fully delivering it, and the scientific and emergency-management communities have spent the decade since trying to make sure the lessons from that near-miss are not wasted.