Low-latitute subduction of marine sediment and deep carbon recycling

Carbon is one of the most pivotal elements on Earth, whose behavior exerts profound influences on the operation of the global climate system, the origin and evolution of life, and the formation of fossil energy resources. Over 99% of Earth’s carbon resides within solid reservoirs, including sediments, crust, mantle, and core. Less than 1% is distributed among fluid reservoirs, encompassing the biosphere, oceans, and atmosphere. Oceanic plate subduction transports substantial shallow-sourced carbon-bearing materials, including sediments, altered oceanic crust, and serpentinized peridotite, into the deep mantle. Some of this subducted carbon is returned to the hydrosphere or atmosphere via mantle wedge melting and arc volcanism, wheras the rest is either stored within the lithosphere or conveyed into the deeper convective mantle. Therefore, oceanic plate subduction plays a critical role in the exchange between solid and fluid carbon reservoirs. Marine sediments overlying the oceanic crust, which are characterized by low density, weak rheological strength, and enrichment of incompatible elements, constitute a major solid carbon reservoir. This study synthesizes recent advancements and unresolved issues in our understanding of deep carbon cycling within subduction zones, including decarbonization mechanisms, the efficiency of carbon release, numerical modeling approaches, and associated climatic feedbacks. Analysis of modern subduction zone sediments reveals that trench sediments in low-latitude regions exhibit elevated carbonate content, potentially attributable to warmer seawater temperatures and enhanced pelagic organism productivity that facilitate carbonate burial in these regions. Only a fraction of subducted carbon (~30%) is liberated to the atmosphere via volcanic degassing after sediments enter the trench. Following subduction, decarbonation of carbonaceous sediments occurs through metamorphic reactions, dissolution, partial melting, and diapirism of carbon-bearing material, though the relative contributions and efficiencies of these mechanisms remain debated. The decarbonation style and efficiency are primarily controlled by the protolith composition of subducted sediments and thermal structure of subduction zones. In addition, the regulatory role of water in decarbonation processes within subduction zones is also crucial. On the one hand, water addition substantially lowers the reaction temperature for carbonate decomposition. Typically initiated above 700°C, the decarbonation reaction can occur at 400–600°C in a hydrous environment, thus enhancing the reaction efficiency. On the other hand, water facilitates the formation of supercritical CO2–H2O fluids or carbonate melts, which serve as vital carriers for carbon transfer from the slab to the mantle wedge or the forearc crust. Numerical modeling results indicate that sediment-derived melts generated by slab heating migrate buoyantly through the mantle wedge to form diapiric structures, a critical mechanism of slab decarbonation. Plate reconstructions show that the Neo-Tethyan subduction zone persisted in low-latitude regions over prolonged geological timescales, facilitating the widespread deposition of carbonate-rich sediments due to elevated pelagic productivity and sedimentation rates near the equator. We suggest that intense magmatism induced by the subduction of these carbon-enriched sediments during the Neo-Tethyan closure may have significantly contributed to Cretaceous-Paleogene hothouse climate. Future research priorities may include the (1) development high-resolution plate reconstruction models to better constrain sediment thickness distributions over geologic time and their linkages to long-term climate change; (2) integration of experiment rock physics results with dense-array magnetotelluric and seismic observations to better image deep carbon storage within subduction zones; and (3) advancement of numerical simulation methods based on two-phase flow dynamics, enabling high-resolution modeling of sediment subduction. This will elucidate the migration and enrichment patterns of carbon-bearing materials released from the subducting slab.