The average Phanerozoic CO2 degassing flux estimated from the O-isotopic composition of seawater

Abstract

The degassing rate of CO2 from the solid Earth, and matching burial rate, is critical to understanding the long-term carbon and oxygen cycles and the maintenance of a habitable climate. However, estimates of modern CO2 degassing rates vary by more than a factor of four and uncertainties are even larger further back in Earth history. Both the crustal C-inventory and C-isotope mass balance suggests that over long timescales ∼70-90% of the CO2 degassed from the solid Earth is drawn-down through silicate weathering reactions (subaerial and submarine). These reactions involve dissolution of silicate minerals, that typically formed at relatively high temperatures, and formation of silicate and carbonate minerals in low temperature aqueous environments. These new low temperature minerals have much higher δ18O than their precursors; thus the rate of degassing of CO2 is tied to the rate of removal of materials with high δ18O from the hydrosphere into weathering products. Maintaining a near-constant O-isotopic composition of the ocean over the Phanerozoic means that this weathering sink of isotopically heavy O must have equaled the source of isotopically heavy O to the ocean. This isotopically heavy oxygen largely comes from high-temperature hydrothermal vent fluids at mid-ocean ridges. New O-isotope data for the previously under-sampled lower oceanic crust from fast-spreading ridges, along with a compilation of published O-isotopic compositions of other sections of oceanic crust altered at high-temperatures, allows the magnitude of this high δ18O flux to be quantified. Modelling the coupled O-isotope and CO2 mass balance constrains suggests that the average Phanerozoic solid Earth CO2 degassing rate was ∼3.4 Tmolyr−1 (10th to 90th percentile of 1.8-5.3 Tmolyr−1). This is substantially lower than has been suggested by some recent studies, and is at or below the lower end of the range of previous estimates, with wide ranging implications for our understanding of the long term C cycle.