The death of supercontinents brings diamonds to the surface

Deep mantle waves trigger mysterious kimberlite volcanoes

by Paul Voosen

Forged under extreme temperatures and pressures more than 150 kilometers down in the mantle, diamonds ride rockets to reach Earth’s surface: narrow pipes of magma called kimberlite that can erupt at the speed of sound. Strangely, most kimberlite pipes are found in the quiet, ancient interiors of continents. They are far from where most other eruptions occur: at the edges of tectonic plates and near mantle plumes, broad upwellings that form volcanic hot spots such as Hawaii or Yellowstone. “How on Earth did these get here?” asks Thomas Gernon, a geologist at the University of Southampton. “It was an elephant in the room that no one had a good explanation for.”

Now, Gernon and his colleagues believe they do. They say the timing and location of these diamond-bearing eruptions suggest they are aftereffects of the breakup of supercontinents, which causes whirling turbulence in the viscous mantle rock below. Like slow-motion tidal waves of rock, the researchers say, these swells ripple beneath the continents, traveling hundreds of kilometers over the course of millions of years—and occasionally triggering kimberlite eruptions. “Kimberlites seem to be responding to rhythms of supercontinents,” Gernon says.

The finding, published today in Nature, is about more than diamonds and kimberlites, says Folarin Kolawole, a structural geologist at Columbia University who is unaffiliated with the study. It suggests that tectonic action near Earth’s surface can influence the behavior of the mantle on a broader scale than once thought. It also indicates that the underground waves keep the margins of newly divided continents volcanically active far longer than expected, possibly explaining other volcanic rocks that had previously been chalked up to mantle plumes. “This gives us a way forward—a hypothesis to test,” Kolawole says.

SIGN UP FOR THE SCIENCEADVISER NEWSLETTER

The latest news, commentary, and research, free to your inbox dailySIGN UP

Named after a mine in Kimberley, South Africa, that produced most of the world’s diamonds in the late 1800s, kimberlite forms when mantle rocks melt into a dense magma rich in magnesium, water, and carbon dioxide. The CO₂ and water create bubbles of gas that drive a narrow plug of magma toward the surface like the cork popping from a bottle of champagne.

At depth, the pipes are just a few meters across, but they expand into carrot-shaped cones when they burst out, leaving craters hundreds of meters across. “It’s going to be quite catastrophic when it reaches the Earth’s surface,” says Emma Humphreys-Williams, a geochemist at the Natural History Museum in London. Along the way, the magma picks up hitchhikers from the upper mantle and continental crust, including diamonds, that are rich in clues to the behavior of the deep Earth. “There’s a whole element of Earth history that would be lost if we didn’t have kimberlites,” says Kelly Russell at the University of British Columbia.

Although Earth is dotted with thousands of known kimberlite pipes, not one has been known to erupt in recorded human history. The vast majority are hundreds of millions of years old. “All of us volcanologists would be willing to chip in and have a new eruption of kimberlites,” Russell says.

During COVID-19 lockdowns, Gernon found himself returning to the question of why these eruptions only occur in cratons—the old, cold interiors of continents. Although some researchers believed mantle plumes could be the source, many kimberlite deposits don’t match known hot spot tracks. In addition, geochemical analysis of isotopes in mantle plume rocks suggest they originate in the lower mantle, whereas recent analyses of kimberlites point to a shallower origin.

Gernon and his co-authors noticed how the timing of kimberlite eruptions did seem to match up with landmark events in plate tectonics. They reconstructed the movement of the continental plates over the past 500 million years, comparing the rate of continental rifting with the bursts of kimberlite formation. The analysis showed kimberlite eruptions seemed to peak, on average, 26 million years after a continental breakup. “That was curious and jaw dropping,” Gernon says.

They then zoomed in on the geologic history of kimberlite deposits in southern Africa and South America, which formed after the breakup of the Gondwana supercontinent 120 million years ago, and those that formed in North America after the earlier crack up of Pangaea. The kimberlite volcanoes popped off progressively farther from the rift over time, the clusters shifting some 20 kilometers every million years.

Gernon and his colleagues think they know what drives this migration. As continents split apart, hot mantle rocks well up to fill the gap. But they cool off and sink as they rub up against the cold continental sides of the gap, creating whirling convective patterns. Computer modeling shows these vortices travel along the keels of the continents, stripping their mushy roots and creating a rock mix perfect for melting into kimberlite. The simulations show the waves crawling along at a pace that matches the propagation of kimberlites: about one-millionth of a snail’s pace. “It’s a really clever idea,” says Jeroen van Hunen, a geodynamicist at Durham University. “It makes perfect sense.”

It’s also a big claim, as the powerful waves would strip some 40 kilometers of rock from the base of continents. But some other evidence seems to support it. For example, not only do kimberlites migrate out from rifts over time, but their mix of isotopes also shifts, from patterns that resemble a mantle-crust mix where the wave first breaks to a more uniform upper mantle composition as the wave dies out. And in the Kaapvaal craton of southern Africa, for example, the continent saw several kilometers of uplift around the same time as the kimberlite eruptions. The uplift suggests the wave was underfoot at the time, stripping off the continent’s undercarriage and allowing it to rise like a hot air balloon shedding its ballast. “Combined, this evidence is really compelling,” Gernon says.

It’s unlikely the team has found the single cause for kimberlites, given how noisy the data are, and how much Earth’s quirks vary from place to place, says Karen Smit, a geochemist at the University of the Witwatersrand. “It’s a model that makes sense. I just don’t know if that correlation exists globally.” But Kolawole says the study is likely to prompt a surge of work in regions such as the Gulf of California or the Red Sea, where incipient rifting might be creating the deep waves, which seismic observations could reveal. The theory might also explain some volcanic deposits that were previously attributed to plumes, he adds.

The greatest interest in the study may come from commercial diamond miners, Gernon says. In theory, it could help predict the location of undiscovered kimberlites, he says. “You should be able to pinpoint, roughly, the sweet spot for diamonds.”