Neogene changes in land surface reactivity and implications for Earth system sensitivity to carbon cycle perturbations

Long-term cooling, pCO₂ decline, and the establishment of permanent, polar ice sheets in the Neogene has frequently been attributed to increased uplift and erosion of mountains and consequent increases in silicate weathering, which removes atmospheric CO₂. However, geological records of erosion rates are potentially subject to averaging biases and the magnitude of the increase in weathering fluxes, and even its existence, remain debated. Moreover, a weathering increase scaled to the hypothesized erosional increase would have removed nearly all carbon from the atmosphere, leading to proposals of compensatory carbon fluxes in order to preserve carbon cycle mass balance. In contrast, increasing land surface reactivity—resulting from greater fresh mineral surface area or an increase in the supply of reactive minerals—rather than an increase in the weathering flux, has been proposed to reconcile these disparate views. We develop a parsimonious carbon cycle model that tracks two weathering-sensitive isotopic tracers (stable ⁷Li/⁶Li and cosmogenic ¹⁰Be/⁹Be) to show that an increase in land surface reactivity is necessary to simultaneously decrease atmospheric CO₂, increase seawater ⁷Li/⁶Li, and retain constant seawater ¹⁰Be/⁹Be since 16 Ma. We find that the global silicate weathering flux remained constant, even as the global silicate weathering intensity—the fraction of the total denudation flux derived from silicate weathering—decreased, sustained by an increase in erosion. Thus, long-term cooling during the Neogene reflects a change in the partitioning of denudation into weathering and erosion. Variable partitioning of denudation and consequent changes in silicate weathering intensity reconcile marine isotope and erosion records with the need to maintain mass balance in the carbon cycle and without increases in the silicate weathering flux. These changes in land surface reactivity through time suggest that the Earth system’s response to carbon cycle perturbations is not constant and that today’s Earth can more efficiently remove excess carbon than during analogous perturbations observed in the geologic record.