Diamond digs
If you want to find diamonds, look for mantle plumes. That is the main message of research published in this week’s Nature.Mantle plumes are huge masses of hot, buoyant rock that periodically rise from the depths of the Earth, almost 3,000km down at the boundary between the iron core and the hot rocks that make up the mantle. Looking like enormous jellyfish, with a broad head narrowing to a thin tail, these mantle plumes slowly travel to the underside of the Earth’s crust, where they proceed to burn their way through.
Depending on the size of the plume, the end result can be the formation of individual volcanoes or the relatively sudden release of huge amounts of molten lava that cover vast areas of the Earth, producing what are now rather boringly known as large igneous provinces (LIPs). Although such ‘hot spot’ volcanoes are still being formed (such as the undersea volcanoes that comprise the Hawaiian Islands), the most recent LIPs are millions of years old. This is fortunate, as the great molten upheavals that produced some of the largest LIPs such as the Deccan Traps in India and the Serbian Traps in Russia have been linked with mass extinctions.
But now an international team of scientists led by Kevin Burke from the University of Houston in Texas has uncovered evidence that mantle plumes may also be responsible for producing diamonds and delivering them to the surface of the Earth.
Mantle plumes were first postulated over 40 years ago to explain the existence of volcanoes in the middle of tectonic plates (most volcanoes form at the edges of plates). For much of this time, the concept of mantle plumes proved highly controversial, with many scientists doubting whether they really existed. Over the past few years, however, by conducting sensitive seismic studies of the Earth’s interior, scientists such as Burke have managed to detect them.
Furthermore, by combining this seismic evidence with models of how the surface of the Earth has changed over the past 300 million years, Burke and his colleagues have traced the mantle plumes that produced hot spot volcanoes and LIPs over that time back to their origins at the core-mantle boundary. In particular, they found that mantle plumes tend to arise at the edges of two vast regions of the core-mantle boundary known as large low-shear-wave velocity provinces (LLSVPs). Although little understood, the unique seismic signatures of these regions suggest they have a different chemical composition to the surrounding mantle.
Now, Burke and his team have gone one step further and matched up known deposits of an igneous rock known as kimberlite with the edges of these two LLSVPs. Named after the town of Kimberly in South Africa, kimberlite is the rock that houses diamonds. It forms deep within the Earths’ mantle, at least 150km below the surface, where the temperatures and pressures are great enough to force carbon atoms together into the regular lattice of diamond. Indeed, kimberlite probably forms at greater depths than any other form of igneous rock.
Burke and his team have now found that around 80% of known kimberlite rocks erupted onto the surface of the Earth at points located directly over the edges of LLSVPs. This strongly suggests that not only did the heat from mantle plumes stimulate the formation of kimberlite and diamonds, but that the rising plumes then brought kimberlite to the surface.
This also explains why certain regions of Africa, including South Africa and the Congo, are comparatively rich in diamonds. Over the past 300 million years, Africa, whether on its own or part of the supercontinent known as Gondwana, has drifted north over the edge of one of the LLSVPs. As a consequence, its underside has repeatedly been hit by kimberlite-carrying mantle plumes.
This work also reveals that these two LLSVPs have remained in a fixed position for at least the past 300 millions years, while the mantle and the Earth’s surface has moved above them. Burke thinks that the LLSVPs may have remained in the same position for much of the Earth’s history, although most of the evidence for this has been erased by the slow churning of the Earth.
The alternative possibility, as proposed by David Evans at Yale University in New Haven, Connecticut, is that LLSVPs form along with each new Earth-spanning supercontinent. So the two current LLSVPs formed along with Pangea, which started breaking up around 180 million years ago, and new LLSVPs will form with the next supercontinent, which is already in the process of coming together.
Mantle plume activity seems to be associated with the break up of supercontinents. Most kimberlite rocks were delivered to the surface of the Earth between 70 and 120 million years ago and there have been no kimberlite eruptions in recorded history. So not only does this work indicate where existing deposits of diamonds are most likely to be found, but it also suggests that deliveries should begin again in a few hundred million years.
