Carbon Sinking Into Earth’s Mantle Leaves a Seismic Fingerprint 560 Kilometers Down

Deep beneath the ocean floors, slabs of old seafloor sink slowly into Earth’s interior, carrying with them far more than cold rock. Locked within their minerals are carbonates, chemical compounds that record millions of years of interaction between the oceans, the atmosphere, and the solid Earth. A new study published in Nature Communications suggests that these humble carbon-bearing sediments are not passive passengers on their descent into the mantle. Instead, they actively transform the mineralogy of the subducted crust in ways powerful enough to reshape one of the most enigmatic features of the planet’s deep seismic architecture: the splitting of the 520-kilometer discontinuity.

The mantle transition zone, lying between roughly 410 and 660 kilometers depth, is defined by abrupt changes in seismic wave speeds caused by pressure-driven mineral transformations. Olivine, the dominant mineral of the upper mantle, converts to wadsleyite near 410 kilometers and to ringwoodite near 520 kilometers, before collapsing into bridgmanite and ferropericlase at the 660-kilometer boundary. These boundaries normally appear as sharp, horizontal reflectors in seismic records. Yet seismologists have repeatedly observed something strange: beneath certain subduction zones and mantle plumes, the 520-kilometer discontinuity appears split, with a secondary reflector emerging near 560 kilometers depth. This sporadic feature has long resisted a satisfying explanation.

One leading hypothesis attributed the 560-kilometer reflector to the exsolution of davemaoite, a high-pressure calcium silicate phase with the formula CaSiO3. In principle, if a calcium-rich rock reaches the depths where davemaoite becomes stable, the sudden appearance of this mineral should create a detectable impedance contrast, a jump in the resistance that seismic shear waves encounter as they pass through the layer. The problem, as the researchers behind the new study recognized, is quantitative. Standard shear-wave properties of davemaoite, when applied to plausible rock compositions, simply failed to generate the 2 to 4 percent impedance contrast that seismic observations demand. The mineral alone, in ordinary mantle compositions, was not abundant enough or contrastive enough to explain the signal.

To resolve this discrepancy, a team at Ehime University turned to large-volume multianvil apparatus, machines capable of squeezing and heating tiny samples to the extreme conditions of the deep transition zone. Their experiments, conducted at pressures near 20 gigapascals and temperatures between 1,200 and 1,600 degrees Celsius, simulated the environment experienced by oceanic crust that has been altered by carbonates before and during subduction. The compositions they tested mimicked carbonate-altered oceanic crust, the kind of material that forms when seawater reacts with basaltic rocks at the seafloor, precipitating calcium carbonate into cracks and veins, and that then rides a descending slab into the mantle.

The results were striking. The addition of carbonate did not merely sit inertly within the rock. Instead, it drove a vigorous chemical exchange, with calcium and magnesium swapping positions between carbonate and silicate phases as the sample equilibrated at depth. Under iron-rich conditions in particular, this exchange dramatically increased the amount of calcium incorporated into the silicate minerals of the crust. The consequence was a profound boost in davemaoite exsolution. When the calcium-enriched silicates reached the pressure range where davemaoite becomes stable, the mineral precipitated in remarkable abundance, reaching concentrations of 12 to 33 volume percent near 560 kilometers depth. That is far more davemaoite than standard mantle compositions could ever produce, and critically, it is enough to account for the observed 2 to 4 percent impedance contrast recorded in seismic data.

This finding reframes the role of carbon in the deep Earth. Scientists have long known that subduction is the primary conduit through which surface carbon returns to the mantle, completing a cycle that regulates atmospheric carbon dioxide over geological timescales and influences the melting behavior, redox state, and volatile content of the deep interior. What the new experiments demonstrate is that this carbon is not chemically quiet during its descent. By catalyzing calcium-magnesium exchange, carbonates restructure the phase equilibria of the subducted crust itself, changing which minerals form, when they form, and in what quantities. The seismic consequences are direct and measurable: a chemical stratification reinforced by carbon, expressed in the seismic record as a persistent reflector where none should exist in a chemically homogeneous mantle.

The mechanism also offers an explanation for the geographic pattern of the observations. The split 520-kilometer discontinuity appears preferentially beneath subduction slabs and beneath hot mantle plumes, and the new study provides a coherent narrative for both settings. In subduction zones, carbonate-altered oceanic crust sinks into the transition zone, becomes calcium-enriched through the experimentally documented exchange process, and exsolves davemaoite at depth, producing the anomalous reflector. In plume settings, the story inverts: upwelling mantle plumes can carry ancient slabs of recycled, carbonate-altered oceanic crust back toward the surface. As this calcium-enriched material rises through the 560-kilometer depth range, it retains its davemaoite-rich character and continues to exhibit the same 2 to 4 percent impedance signature, explaining why persistent 560-kilometer reflectors are detected beneath hot thermal regions far from any active subduction.

In this sense, the 560-kilometer discontinuity becomes something like a geological archive written in seismic waves. Every detection of the split reflector marks the presence of rock that once lay at the surface, absorbed carbon from the oceans, and made a round trip through the deepest reaches of the transition zone. Because the chemical modification is durable, the signature can survive for hundreds of millions of years, offering seismologists a rare window into the long-term history of deep carbon cycling. Regions of the mantle that appear seismically anomalous may therefore be cataloging the planet’s carbon past, recording where slabs dove, where carbon was delivered, and where plumes have dredged recycled crust back upward.

The study’s experimental approach is what gives the claim its force. Rather than extrapolating from mineral physics of idealized compositions, the researchers directly reproduced the pressure-temperature path of carbonated crust and measured the resulting phase assemblages. The volume fractions of davemaoite they documented, 12 to 33 volume percent, fall squarely within the range needed to reconcile the seismic observations, closing a gap that had persisted between the exsolution hypothesis and the actual impedance contrasts measured in the field. The identification of iron-rich conditions as a key enabler of calcium incorporation adds an important compositional control, suggesting that the visibility of the effect depends on the detailed chemistry of the subducted material, which may help explain why the split discontinuity is observed in some locations and not others.

More broadly, the work underscores a theme increasingly emphasized in deep Earth research: the interior of the planet is not a chemically bland, well-mixed reservoir, but a heterogeneous archive whose structure is actively shaped by surface processes. Carbonates raining down on slabs today will, hundreds of millions of years from now, leave their mark on the seismic images of tomorrow’s mantle. The boundary near 560 kilometers depth, once a curiosity of seismology, now stands as evidence that the carbon cycle does not stop at the base of the crust. It reaches deep into the transition zone, rewrites the mineralogy of everything it touches, and broadcasts its presence upward in the language of seismic waves, waiting for instruments at the surface to translate it.

Subject of Research: The role of subducted carbonates in driving davemaoite exsolution and the seismic splitting of the 520-km mantle discontinuity

Article Title: Subducted carbonates drive deep carbon cycling and reshape mantle discontinuity structure

Article References: Subducted carbonates drive deep carbon cycling and reshape mantle discontinuity structure. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: deep carbon cycle, subduction, mantle transition zone, 520-km discontinuity, davemaoite, carbonates, oceanic crust, seismic discontinuity, high-pressure experiments, mantle plumes, calcium-magnesium exchange, Nature Communications