Long-term stability of global erosion rates and weathering during late-Cenozoic cooling.

Over geologic timescales, CO(2) is emitted from the Earth's interior and is removed from the atmosphere by silicate rock weathering and organic carbon burial. This balance is thought to have stabilized greenhouse conditions within a range that ensured habitable conditions. Changes in this balance have been attributed to changes in topographic relief, where varying rates of continental rock weathering and erosion are superimposed on fluctuations in organic carbon burial. Geological strata provide an indirect yet imperfectly preserved record of this change through changing rates of sedimentation. Widespread observations of a recent (0-5-Myr) fourfold increase in global sedimentation rates require a global mechanism to explain them. Accelerated uplift and global cooling have been given as possible causes, but because of the links between rates of erosion and the correlated rate of weathering, an increase in the drawdown of CO(2) that is predicted to follow may be the cause of global climate change instead. However, globally, rates of uplift cannot increase everywhere in the way that apparent sedimentation rates do. Moreover, proxy records of past atmospheric CO(2) provide no evidence for this large reduction in recent CO(2) concentrations. Here we question whether this increase in global weathering and erosion actually occurred and whether the apparent increase in the sedimentation rate is due to observational biases in the sedimentary record. As evidence, we recast the ocean dissolved (10)Be/(9)Be isotope system as a weathering proxy spanning the past approximately 12 Myr (ref. 14). This proxy indicates stable weathering fluxes during the late-Cenozoic era. The sum of these observations shows neither clear evidence for increased erosion nor clear evidence for a pulse in weathered material to the ocean. We conclude that processes different from an increase in denudation caused Cenozoic global cooling, and that global cooling had no profound effect on spatially and temporally averaged weathering rates.