Rogue Waves Explained: What Causes Monster Waves, and How Often They Strike

Jump to comments

On the afternoon of New Year’s Day 1995, a laser pointed straight down from an unmanned gas platform in the North Sea logged a wave that, by the textbooks of the time, should not have existed. The storm around the Draupner E platform was producing waves of about 12 meters (39 feet). Then one arrived 25.6 meters (84 feet) from trough to crest, with its peak 18.5 meters above the calm-water line. For centuries sailors had described walls of water that appeared out of nowhere, and for centuries scientists had filed those stories under exaggeration. The Draupner record ended that argument in one 20-minute data file.

The short answer: a rogue wave is a single wave more than about twice as tall as the waves around it. Most form when several wave trains cross and their crests briefly line up, a process called linear superposition, with the sea’s own nonlinear physics and ocean currents adding height in some cases. They are rare for any one wave but common across the whole ocean: a buoy study of more than a billion waves found about 100,000 that met the rogue definition. Here is what counts as a rogue wave, the three ways the ocean builds one, how often they happen, the ships that met them, whether they can be forecast, and what the beach version means for anyone standing near the water.

What counts as a rogue wave?

Oceanographers describe a stormy sea with one number, the significant wave height, which is roughly the average height of the highest third of the waves. It is close to what an experienced sailor would report by eye. A rogue wave is defined against that number, not in absolute feet. The National Ocean Service calls rogues, or "extreme storm waves," waves "greater than twice the size of surrounding waves" that often come from a different direction than the prevailing wind and swell. Researchers use stricter versions: a trough-to-crest height above 2 or 2.2 times the significant wave height, or a crest that rises more than 1.25 times the significant wave height above the still-water level.

That makes "rogue" a relative word. A 2-meter wave in a 1-meter sea is a rogue wave. So was the 25.6-meter Draupner wave in a 12-meter sea. What makes rogues dangerous is less their raw size than the surprise: they are steep-sided, often with an unusually deep trough in front, and they arrive among waves that a ship, platform or beachgoer has already judged to be manageable.

Rogue waves are also not tsunamis. A tsunami is a long, fast pulse pushed out by an earthquake, landslide or eruption, like the waves that crossed the Pacific after the 1960 Valdivia earthquake. It can be barely noticeable on the open ocean and only towers when it reaches shallow water. A rogue wave is an ordinary wind-driven wave that briefly becomes extraordinary, and it is usually gone again within a minute.

The wave that made scientists believe

Draupner E had been instrumented for a mundane reason. It was the first platform built on a bucket foundation, so its operator, Statoil, installed a downward-looking laser wave sensor to check how the foundation handled its first winter. In a short account of the event, Statoil engineer Sverre Haver described a deep low centered over southern Sweden driving strong northerly winds across the North Sea, reaching hurricane force in places that afternoon. The significant wave height at the platform reached 11 to 12 meters soon after noon. In a sea like that, he wrote, the biggest wave should be about 20 meters, "occasionally" 10 to 15 percent higher, with a crest about 15 meters above the still-water line. The recorded crest was 18.5 meters.

The wave did minor damage to equipment under the deck, and the platform’s sparse frame of legs meant the laser saw something close to an undisturbed sea. That is why the record mattered so much. It was not a captain’s estimate from a pitching bridge. It was a clean instrument trace, and the European Centre for Medium-Range Weather Forecasts later summed it up simply: the measurement "confirmed the existence of giant rogue waves, which had previously been reported anecdotally by sailors."

Other evidence followed quickly. The European Space Agency reported in 2004 that radar on platforms in the North Sea’s Goma oilfield had logged 466 rogue-wave encounters in 12 years, at a time when standard statistics said such large departures from the surrounding sea should happen once every 10,000 years. ESA’s own satellite search, part of an EU project called MaxWave, scanned about 30,000 radar snapshots of the ocean covering just three weeks in 2001 and found more than ten individual waves taller than 25 meters.

Three ways the ocean builds a monster wave

Ocean waves travel in groups, and each group is a mix of many waves with slightly different lengths, speeds and directions. Most of the time those components partly cancel and partly reinforce one another, producing the familiar irregular sea. A rogue wave is what happens when far more of them than usual pile up in the same place at the same moment. Scientists have found three main ways that can happen.

1. Crossing and stacking

The simplest explanation is also the oldest. When two or more wave systems cross, say a fresh wind sea and a long swell from a distant storm, their crests occasionally coincide. The National Ocean Service describes swells "crossing the ocean at different speeds and directions," whose crests, troughs and lengths "sometimes coincide and reinforce each other," forming towering waves that quickly disappear. Physicists call this linear superposition, and it needs no exotic physics at all: it is simple addition.

For years many researchers suspected that superposition alone was too feeble to make the biggest rogues. Big data has pushed the field back toward it. In 2021 and again in 2023, Dion Häfner and colleagues at the University of Copenhagen’s Niels Bohr Institute and the University of Victoria fed more than a billion buoy-recorded waves into machine-learning models. The strongest predictor of rogue waves, in every depth and location they studied, was a measure of how well the crests and troughs in a sea line up, called crest-trough correlation. It explained about a tenfold variation in risk. Their conclusion, first published in 2021 and refined in PNAS in November 2023, was that linear superposition in seas with a narrow spread of wave periods is the main pathway to "everyday" rogue waves, with nonlinear effects as a smaller correction.

2. The wave that steals from its neighbors

The second mechanism is stranger. Steep waves of nearly the same length are unstable: in 1967 T. Brooke Benjamin and J. E. Feir showed that a smooth train of such waves breaks up into groups, with energy drifting from some waves into others. In the most extreme version, one wave grows to several times the height of the background for a moment and then shrinks back, leaving no trace. The mathematical form of that lone giant is called the Peregrine soliton, after Howell Peregrine of the University of Bristol, who described it in 1983. In 2011 Amin Chabchoub and colleagues produced one in a 15-meter water tank, a peak about three times the height of the waves around it, reported in Physical Review Letters.

This process, called modulational instability, is easy to trigger in a narrow laboratory tank, where waves all move in one direction. In the real ocean, waves spread out in direction, and that spreading weakens it. For the Draupner storm, Oxford researchers Thomas Adcock and Paul Taylor noted that the standard instability index, about 0.55 on paper, falls to roughly 0.07 once the platform’s 70-meter water depth is allowed for, and lower still with directional spreading. Most researchers now see it as a contributor in special conditions, not the main engine.

3. Currents that act like a lens

The third mechanism is geographic. When storm waves run into a strong current flowing against them, the waves shorten and steepen, and the current can bend and concentrate their energy. Susanne Lehner, a radar scientist on the MaxWave project, compared the effect to an optical lens. The classic example is the Agulhas Current off South Africa’s east coast, long notorious among mariners, with the Gulf Stream another hot spot. NOAA’s own rogue-wave page is illustrated with a wave estimated at 18.3 meters (60 feet) that hit a ship in the Gulf Stream off Charleston, South Carolina, when surface winds were only about 15 knots.

Crossing seas can also defeat the normal limit on wave height. A wave usually breaks once it gets too steep, which caps its size. In 2019 a team led by Mark McAllister and Ton van den Bremer of the University of Oxford, working with Sam Draycott of the University of Edinburgh, recreated the Draupner wave in Edinburgh’s FloWave basin, a circular tank with wavemakers all the way around. They could only reproduce it when two wave groups crossed at about 120 degrees. At that angle the crest did not topple forward and stop growing. It broke upward, as a vertical jet, and kept climbing.

The recreated wave looked so much like Katsushika Hokusai’s woodblock print "The Great Wave off Kanagawa" that the team highlighted the resemblance in the University of Oxford’s announcement. McAllister called the Draupner measurement "a seminal observation," one that shifted freak waves "from mere folklore to a credible real-world phenomenon."

How common are rogue waves?

Rare for any single wave, routine for the ocean as a whole. The Copenhagen and Victoria team’s dataset drew on buoys at 158 locations around the U.S. coasts and overseas territories, about 700 years of combined records. Their pre-filtered catalog, described in Scientific Reports, held about 1.5 billion individual waves, of which about 100,000 were taller than twice the significant wave height. "This is equivalent around 1 monster wave occurring every day at any random location in the ocean," co-author Johannes Gemmrich said in the University of Copenhagen’s announcement, adding that they "aren’t all monster waves of extreme size."

Those figures allow a rough check of the odds. One hundred thousand rogue waves out of 1.5 billion is about one wave in 15,000. The classic statistics for a random sea, the Rayleigh distribution, put the chance of a wave topping twice the significant height at about one in 3,000, and the researchers found that in most conditions that textbook figure is actually an upper bound for real seas. With a typical storm-wave period of 10 seconds, one in 15,000 works out to roughly one rogue wave every 40 hours at a given spot. A ship spending five days in rough water meets more than 40,000 waves. In that case the question is not whether it will meet a rogue wave, but how big the sea is when it does.

The question is not whether a ship will meet a rogue wave, but how big the sea is when it does.

The most extreme rogue wave on record, measured against its surroundings, was not in a hurricane. On November 17, 2020, a small buoy operated by MarineLabs about 7 kilometers off Ucluelet on Vancouver Island, in 45 meters of water, recorded a wave 17.6 meters high in a sea whose significant height was about 6 meters. Its crest rose 11.96 meters, and the waves just before and after it had crests of only 3.27 and 4.13 meters. Gemmrich and Leah Cicon of the University of Victoria reported in Scientific Reports that its normalized height, 2.9 times the significant wave height, was likely the largest ever recorded, and that second-order theory put such an event at once in 1,300 years.

Proportionally, the Ucluelet wave is likely the most extreme rogue wave ever recorded. Only a few rogue waves in high sea states have been observed directly, and nothing of this magnitude. (Johannes Gemmrich, University of Victoria, in a MarineLabs announcement, February 2022)

For context, the biggest seas measured are larger still. The World Meteorological Organization certified a significant wave height of 19 meters (62.3 feet), recorded by a buoy between Iceland and the United Kingdom on February 4, 2013, as the highest ever measured by a buoy, The Guardian reported. A rogue wave in a sea like that would be close to 40 meters, though no one has measured one.

The ships that met them

Queen Elizabeth 2, September 1995. Eight months after Draupner, the liner QE2 was heading for New York through the tail of Hurricane Luis off Newfoundland when the bridge crew saw a wave approaching at about bridge height, roughly 95 feet (29 meters) above the waterline. "It looked as if we were going straight into the white cliffs of Dover," Captain Ronald Warwick told reporters, The Herald reported, calling it the biggest he had seen in 38 years at sea. The wave broke over the bow and damaged rails, and the ship reached New York eight hours late.

Bremen and Caledonian Star, 2001. In one week between February and March 2001, two expedition cruise ships in the South Atlantic had their bridge windows smashed by waves ESA put at about 30 meters. The Bremen drifted without navigation or propulsion for about two hours. The two encounters were less than 1,000 kilometers apart, and the three weeks of radar data the MaxWave team searched were picked around them.

Viking Polaris, November 2022. On the night of November 29, 2022, the expedition ship Viking Polaris was crossing the Drake Passage toward Ushuaia, Argentina, when a breaking wave struck its side southeast of Cape Horn. Seven stateroom windows shattered, one passenger died and eight were injured. The Norwegian Safety Investigation Authority found that the force of the wave exceeded what the windows were designed to withstand, that the design rules for windows in that position "yield too low values," and that the crew could not have predicted the risk. It recommended that classification rules be rewritten to account for breaking waves against a ship’s side.

Rogue waves have long been suspected in the loss of ships that vanished in bad weather without a mayday, but in most of those cases no one can know. ESA’s 2004 report noted that severe weather had sunk more than 200 supertankers and container ships longer than 200 meters in the previous two decades. The cause of such losses, one MaxWave scientist complained, was rarely studied in the detail of an air crash and simply recorded as "bad weather."

Can rogue waves be predicted?

Not wave by wave. No forecast can say that a 25-meter wall of water will reach a particular spot at a particular minute. What forecasters can do is flag conditions in which rogue waves are more likely, and that skill has improved quickly. The European Centre for Medium-Range Weather Forecasts has long published an expected maximum wave height in its operational wave model, and a common rule of thumb flags trouble where that maximum exceeds 2.2 times the significant height.

The newer approach builds on the crest-trough finding. In a 2024 paper in the Journal of Physical Oceanography, Leah Cicon, Gemmrich, Häfner and Markus Jochum tested a formula their machine-learning model had distilled from buoy data against ECMWF’s global wave hindcasts. It outscored every other model they tried, while ECMWF’s existing freak-wave indicator did not predict where rogue waves actually occurred. Gemmrich and Cicon had already shown that the crest-trough correlation computed from a standard WAVEWATCH III forecast gave useful warning of the Ucluelet event.

A study published in February 2026 in Ocean Science added a time signal. Looking at rogue waves recorded by buoys, Laura Azevedo and colleagues found that they were usually preceded by a sharp drop in crest-trough correlation below 0.5 followed by a rapid rise above 0.6, along with a narrowing of the wave spectrum. That pattern matched 75 percent of cases in their test. They also found that global models systematically underestimate the most extreme waves because they smooth over the small scales where those waves form. The practical payoff the Copenhagen team pointed to is routing: with about 50,000 cargo ships at sea at any moment, a risk map lets a shipping company steer around the worst combinations of crossing seas and currents.

Are rogue waves getting bigger?

Nobody has enough direct measurements of rogue waves to say whether they are becoming more common. The background they rise from is changing, though. Ian Young and Agustinus Ribal of the University of Melbourne analyzed about 4 billion satellite measurements from 31 satellites between 1985 and 2018 for a 2019 study in Science. In the Southern Ocean, extreme winds rose about 8 percent and extreme wave heights about 30 centimeters, or 5 percent, over the period. Because rogue waves are defined relative to the sea around them, a higher baseline means taller rogues even if their relative frequency stays the same. Young cautioned that it is not yet clear how much of the change is long-term climate change and how much comes from multi-decade cycles.

The beach version: sneaker waves

Most people will never be on a ship in a 12-meter sea, but many stand on beaches where a smaller cousin of the rogue wave kills people every year. The National Weather Service calls them sneaker waves: larger-than-average swells that surge far higher up the sand than the waves before them, often after a lull of 10 to 20 minutes. Some run more than 150 feet (45 meters) up the beach, and the NWS warns that "for much of the West Coast, sneaker waves kill more people than all other weather hazards combined." The physics is the same idea of wave groups stacking, delivered onto steep, cold Pacific Northwest beaches where heavy wet driftwood logs can roll onto anyone sitting on them.

The Weather Service’s advice is short. Watch the water for at least 20 minutes before settling in, the way surfers and lifeguards do. Stay much farther back than seems necessary. Never stand or sit on beach logs. And never turn your back on the ocean.

The bottom line

Rogue waves went from sailors’ tales to measured fact in a single afternoon in 1995, and three decades of buoys, satellites and wave tanks have since explained most of what makes them. The biggest surprise is how ordinary the main recipe is: wave trains crossing and briefly adding up, sometimes helped by a current or by the sea’s own nonlinear focusing. That ordinariness is why they turn up in every ocean, about one in every 15,000 waves in the buoy record, and why forecasters now focus on spotting the sea states that load the dice rather than predicting any single wall of water. For more on the strange things the sea surface can do, see how Fata Morgana mirages make ships float in the sky, and why masts glow with St. Elmo’s fire in storms.

Related on aglena: The Magnus Effect Explained: Why Spinning Balls Curve and Rotor Sails Push Ships (2026)

Further reading