Geochemical consequences of increased Late Cenozoic weathering rates and the global CO2 balance since 100 Ma

Abstract

Large imbalances in the relative net CO2 flux over the last 100 m.y. are obtained from independently derived estimates of CO2 uptake by weathering and organic carbon burial and of CO2 outgassing assumed proportional to the mean mid‐ocean ridge (MOR) spreading and mantle plume production (MPP) rates. Increasing the river flux of Ca, Mg, P, and C by 50% since 30 Ma to parallel the proposed increase in the Sr river flux needed to explain the Sr isotopic evolution of seawater [Richter et al., 1992] has a large impact on the global carbon cycle by reducing the excess net CO2 imbalance in the late Cenozoic. CO2 uptake is calculated by assuming that 19% of the Ca river flux and 60% of the Mg river flux are derived from weathering silicates and that Mg is removed in the oceans by 1:1 molar exchange reactions with calcsilicates. An increase in the Mg river flux combined with an overall decrease in the MOR spreading rate predicts as much as a factor of 2 increase in the seawater Mg concentration since 100 Ma. The net organic carbon burial flux (burial minus weathering) largely reflects changes in the bulk carbonate δ13C record and, to a lesser extent, variations in the net carbonate burial flux and mean δ13C value of the carbonate weathering flux. Organic carbon burial efficiency is markedly less than that of P from the middle middle Miocene to the present because of early diagenetic transfer of organic P to francolite and oxidation of organic carbon during reworking of margin sediments by sea level fluctuations. The large CO2 imbalance of the Late Cretaceous requires significantly less carbonate subduction or greater weathering rates. A CO2 balance can be achieved in the Miocene by decreasing the proposed increase in river fluxes by up to half or by variably increasing the amount of carbonate subducted. Variations in shallow water carbonate burial, and hence the global carbonate record, are currently too poorly known to differentiate between the above possibilities. The decrease in the excess CO2 flux generally parallels the trend to cooler climate inferred from the δ18O record, but periods of large CO2 imbalance generally precede δ18O shifts by several million years.