The Eye of the Sahara: What Really Made the Richat Structure’s Giant Rings

Jump to comments

In the western Sahara, in the middle of Mauritania, there is a pattern in the rock so large that you can’t really see it from the ground. From orbit it looks like a giant bullseye, or an eye staring back up at space: ring inside ring inside ring, about 40 kilometers (25 miles) across. Its formal name is the Richat Structure, or Guelb er Richât. Most people know it by its nickname, the Eye of the Sahara.

The short answer: the Eye of the Sahara is not a meteorite crater, a volcano or the ruins of a lost city. It is a dome of layered rock that was pushed up from below by molten rock about 100 million years ago and has been slowly worn flat ever since. Hard layers of quartzite were left standing as circular ridges and softer layers wore down into valleys, so the inside of the dome is now laid out on the desert floor like the growth rings of a sawn-off tree. Here is how geologists worked that out, why the impact idea lasted so long, and why the Atlantis story that keeps going viral doesn’t hold up.

What you are looking at

The structure sits on the Adrar Plateau in northern Mauritania, near the old caravan town of Ouadane, at about 21 degrees north. NASA’s Earth Observatory and the International Union of Geological Sciences both put it at about 40 kilometers wide. You will also see 45 kilometers (an older NASA figure) and 50 kilometers (Guinness World Records), because the outer rings fade into the plateau and there is no single obvious edge.

Seen in a satellite image, the rings are low ridges called cuestas, separated by sandy or gravelly valleys. Each ridge is a tilted layer of hard rock with a gentle slope on one side and a steeper edge on the other. Japan’s space agency JAXA, describing a 2003 satellite view, says the resistant layers near the center form ridges about 100 meters high and the outer rim reaches about 485 meters above sea level. That is not much relief across a feature 40 kilometers wide. On the ground you see one ridge, then a flat stretch, then another ridge, and the circle only comes together from far overhead.

The local names describe it from the ground and from daily life. Rīšāt means feathers in Hassaniya Arabic. The structure is also called tagense, the word for the round opening of the leather bucket used to draw water from wells.

How a dome turns into a bullseye

The easiest way to picture the Richat is to start with a stack of flat, colored layers, like a layer cake. These rocks were laid down as sand, mud and limy sediment on the floor of a huge, slowly sinking basin, the Taoudeni Basin, starting in the late Proterozoic eon, before about 540 million years ago, and continuing into the Ordovician period, roughly 485 to 443 million years ago.

Step 1: Something pushes up from below

In the middle of the Cretaceous period, around 100 million years ago, a body of molten rock rose underneath the area. It never broke through as a big volcano. Instead it bowed the overlying layers upward into a broad, round blister, a geologic dome. Geologists call the source an alkaline igneous complex, a kind of magma unusually rich in sodium and potassium. The evidence is in the rocks now exposed at the surface: dozens of narrow dikes of carbonatite (an odd igneous rock made mostly of carbonate minerals), a kimberlite plug, and volcanic rocks that geologists read as the roots of small explosion craters. Fission-track dating of apatite in the carbonatites gave about 99 million years, and argon dating of later alteration gave about 98 million.

Step 2: Hot water eats the middle

The doming cracked the rock above the intrusion, and hot, mineral-rich water circulated through the fractures. In the center of the dome that water dissolved limestone and dolomite, much as groundwater carves caves, until the roof collapsed into a jumble of broken rock. That collapse rubble, later cemented with silica, is the Richat’s central breccia: a lens of shattered rock at least 40 meters thick in the middle that thins to a few meters at its edges. In their 2005 paper in Geology, Guillaume Matton, Michel Jébrak and James Lee described it as an "exceptionally well preserved hydrothermal karst infilling" at the summit of the buried complex. As far as they knew, that kind of collapse on top of an alkaline complex had not been described anywhere else, though they suspected it may be more common than anyone realized.

Step 3: Erosion planes it off

For tens of millions of years since, wind and occasional water have stripped the top off the dome. Imagine slicing the top off that layer cake after it has been bent into a hump. Where the knife cuts through, each layer shows up as a ring, with the oldest layers in the middle and the youngest around the outside. That is exactly the order at Richat: late Proterozoic rocks in the center, Ordovician sandstone around the edges, all dipping gently outward at about 10 to 20 degrees.

Step 4: Hard rings stand up, soft rings wear down

Not every layer wears away at the same speed. Quartzite, a sandstone so thoroughly cemented that it breaks through the grains rather than around them, resists erosion and stands up as the circular ridges. Softer shale and siltstone between them wear down into the valleys. Geologists call this differential erosion. It is the same process that left Saudi Arabia’s split Al Naslaa rock balanced on two pedestals, working here on a scale of tens of kilometers instead of a few meters.

A 2024 study in Lithos added a twist. The two dark rings of gabbro, an igneous rock, that circle the center about 3 and 8 kilometers out had long been treated as ring dikes from the same Cretaceous event. El Houssein Abdeina of the University of Nouakchott and colleagues modeled new argon measurements and concluded they are probably much older: flat sheets of magma injected between the sedimentary layers between about 230 and 200 million years ago, during the Central Atlantic Magmatic Province, the vast outpouring of lava that came with the first opening of the Atlantic Ocean. About 100 million years later, the alkaline intrusion domed those flat sheets along with everything else, and erosion cut them into rings. In other words, the Eye of the Sahara records two separate pulses of magma, roughly 100 million years apart.

The crater that wasn’t

A round feature with rings looks a lot like an impact crater, and for a while that was the leading idea. French geologists first described it in the 1930s and 1940s as the "buttonhole" of Richat, and in 1948 the geographer Jacques Richard-Molard suggested it had been pushed up by an intrusion. A 1952 expedition led by the naturalist Théodore Monod logged four circular features in the region: Richat, Aouelloul, Temimichat-Ghallaman and Tenoumer. The other three did turn out to be impact craters, which made Richat look like it belonged to the same family.

1964: researchers led by André Cailleux reported coesite, a form of quartz that forms under the extreme pressure of a meteorite strike, in breccia from the center. That seemed close to proof.

June 4, 1965: on their 12th orbit, Gemini IV astronauts James McDivitt and Ed White photographed the whole structure out the spacecraft window. A NASA report on the mission’s terrain photography says Richat was not on their target list, but the crew had been asked to look for large circular features that might be the roots of impact structures, and it says the structure was of particular interest because of the coesite report. The picture made the Richat famous. It did not settle the argument.

1969: Robert Fudali reported in Science that the supposed coesite was actually barite, an ordinary mineral. The same year he, Robert Dietz and William Cassidy published a paper in the Geological Society of America Bulletin titled, bluntly, "Richat and Semsiyat Domes (Mauritania): Not Astroblemes." Field and lab work found no significant evidence of shock metamorphism, the telltale damage a high-speed collision leaves in rock.

1990s to today: dating work in the 1990s tied the igneous rocks to the mid-Cretaceous. Matton and Jébrak’s papers in 2005 and 2014 described the full intrusion-and-collapse picture, a 2021 gravity and magnetic survey modeled a large magma body beneath the rings, and the 2024 Lithos study pinned down the older age of the gabbro rings. In October 2022 the International Union of Geological Sciences named the Richat one of its first 100 geological heritage sites, calling it "a spectacular example of a magmatic concentric alkaline complex."

It is a case in point that shows serious scientific inquiry can lead to definite answers. (International Union of Geological Sciences, Richat Structure heritage site description)

Is the Eye of the Sahara Atlantis?

No. The claim that the Richat is the lost city of Atlantis has circulated online for years, and in 2018 a YouTube video arguing it went viral enough to be picked up by news sites. Neurologist and skeptic Steven Novella took it apart on his blog that November, and archaeologist Sean Rafferty’s 2025 book Mythologizing the Past points out that, apart from being circular, the Richat is not a very good match at all. The idea rests on one real resemblance and a lot of stretching. Here is how the details compare with what Plato actually wrote.

The rings

In Plato’s dialogue Critias, the god Poseidon surrounds the hill at the center of Atlantis with "alternate zones of sea and land," two of land and three of water, turned "as with a lathe." That is the one real point of contact. But concentric rings are common in nature. The Richat’s rings are tilted rock layers, how many count as "land" or "water" depends on where you decide each one starts, and they are broken and incomplete in places.

The size

Plato gives exact measurements. The central island is 5 stadia across, and the rings of water and land around it are 1, 2, 2, 3 and 3 stadia wide. Add those up and the whole ringed complex is 27 stadia across. A Greek stadion was roughly 180 to 190 meters, so that is about 5 kilometers. Proponents get to a match with the Richat by adding the outer city wall, which Plato puts 50 stadia beyond the outermost ring, for a total of 127 stadia, about 23 to 24 kilometers. Even that falls well short of the Richat’s 40 kilometers, and the wall was a man-made structure on flat ground, not another ring.

The sea

Plato’s Atlantis was an island in the ocean "in front of the straits which are by you called the Pillars of Heracles," with a canal 50 stadia long running from the sea to the outer harbor. The Richat is roughly 500 kilometers from the Atlantic coast and stands hundreds of meters above sea level. Supporters have to assume the ocean once reached it. As Novella notes, there is no evidence that the Atlantic flooded this part of the Sahara in human times.

The ending

In Plato’s Timaeus, Atlantis disappears "in a single day and night of misfortune" and sinks into the sea. The Richat did not sink anywhere. It is a dome of rock that has been eroding in place for tens of millions of years, and its rings are made of rock layers hundreds of millions of years old.

The city

This is the real problem. A large, wealthy city of temples, harbors, bridges and canals would leave walls, foundations, pottery, tools and graves. Archaeologists have surveyed the Richat area for decades and found plenty of human traces, described below, but nothing like a city. Most classicists also read Atlantis as a story Plato invented to make a philosophical point about the ideal state, not as a record of a real place.

People really did live here

The Atlantis story undersells what is actually at the Richat. The structure holds exceptional concentrations of Acheulean stone tools. These are the teardrop-shaped hand axes and cleavers made by early humans, including Homo erectus, over a span of more than a million years. Théodore Monod first documented the sites in 1974. They cluster along dry riverbeds in the outermost valley of the structure, next to outcrops of quartzite that made good raw material, and are mostly absent from the inner rings.

Later people left their marks too. Researchers have recorded Aterian tools made by modern humans between about 145,000 and 29,000 years ago, thousands of stone burial mounds along the ridges, and rock carvings of horsemen with javelins, chariots, cattle and elephants. Radiocarbon dates show that much of the sediment in the valleys was laid down between about 15,000 and 8,000 years ago, during the African Humid Period, when this part of the Sahara had lakes, streams and grassland. The Sahara itself had dried out and greened up many times before that, as the climate swung back and forth. None of that history involves a lost city. It is evidence of something more interesting: people using this landscape, on and off, for hundreds of thousands of years.

Can you visit the Eye of the Sahara?

Yes, but it takes planning. There is no paved road to the structure itself. Visitors usually travel to Atar, the regional hub, which has an airport, then go by four-wheel drive with a local driver and guide to Ouadane and on to the rings. Tour operators put Ouadane about 30 to 50 kilometers from the center of the structure, depending on the route, across unmarked sand and rock. Ouadane is worth the trip on its own: its stone old town is part of the UNESCO World Heritage site of Mauritania’s ancient caravan towns, founded in the 11th and 12th centuries to serve camel caravans crossing the Sahara.

Check your government’s travel advice before you go. As of November 2026, the UK Foreign Office advises against all but essential travel to large parts of Adrar province and lists the town of Ouadane among the exceptions, so where exactly you go matters. The UK warns that insurance can be invalidated by travel against its advice. Many visitors find that the best view of the Eye is still the one from space.

The bottom line

The Eye of the Sahara looks designed, and that is why it keeps going viral. The explanation is ordinary geology working at an extraordinary scale. Molten rock domed up a stack of ancient layers about 100 million years ago, hot water hollowed out and collapsed its core, and erosion sliced the whole thing flat so that every hard layer shows as a ring. It took geologists from the 1930s to the 2020s to work out the details, including a false lead from misidentified coesite and a surprise second age for the gabbro rings. The real story has more in it than the Atlantis version: two pulses of magma 100 million years apart, a vanished cave system in the middle, and stone tools left by people who lived here long before anyone wrote about Atlantis.

Related on aglena: The Sailing Stones of Death Valley: How Paper-Thin Ice Pushes Rocks Across a Dry Lake, When the Mediterranean Dried Up: The Salt Desert Under the Sea and the Flood That Refilled It

Further reading