The early Earth carbon cycle

Marc Hirschmann

University of Minnesota

Because the residence time of carbon in the mantle is >1 Ga, the distribution of carbon on Earth in part reflects in part early events, when fluxes of carbon between the mantle and the exosphere were different from those today. Unfortunately, the early Earth carbon cycle is poorly understood, as petrologic, geochemical, and stratigraphic evidence are not easily reconciled. Early degassing of the planet, as evidenced by Xe isotopes and evidence for early oceans, presumably also created a massive C surface reservoir, but there must have been large-scale ingassing of carbon at some later time, as the present inventory of mantle carbon exceeds that in the exosphere and the H/C ratio of the exosphere is substantially greater than that of the mantle. Did C ingassing occur rapidly in a catastrophic event in the Hadean, or did it occur more recently and gradually? In fact, the surface carbon reservoir, as evidenced by the stratigraphic record, has been growing at least through the Phanerozoic and much of the Proterozoic, which may argue for an early ingassing event. Yet, thermobarometric investigations of Hadean, Archaean and early Proterozoic subduction temperature-depth paths suggest that carbonate would not be retained in subducted lithosphere, as the temperature-depth trajectories are hotter than that required to avoid either decarbonation or carbonate melting. Therefore, a key unresolved question is how and when large-scale carbon ingassing occurred (or alternatively, how the earliest mantle managed to retain much of its carbon). Has modern subduction been able to ingas sufficient amounts of carbon to account for the present inventories? Or perhaps the ingassing occurred early by subduction of reduced carbon rather than carbonate. These questions remain unresolved and provide fertile areas for an initiative aiming to understand deep carbon cycling.