Interactions between carbon dioxide, climate, weathering, and the Antarctic ice sheet in the earliest Oligocene
D. Pollard, L. R. Kump, and J. C. Zachos
Global and Planetary Change (December 2013)
DOI: 10.1016/j.gloplacha.2013.09.012
Abstract A coupled set of models is used to explore the possibility of long-term internal cycles in the CO2–climate–weathering–Antarctic Ice Sheet system. Cycles of this type were found in an earlier study with 0-D box models, and proposed to explain the quasi-periodic oscillations in benthic deep-sea-core records during the Eocene–Oligocene Transition 34 Ma. Here the system is extended using a 3-D Global Climate Model, a 3-D Antarctic ice-sheet model, and two previously published spatially distributed parameterizations of CO2 consumption by silicate weathering. In 6-million year long simulations across an idealized Eocene–Oligocene Transition, no internal cycles are found, and the coupled system just relaxes from the initial state to the final state, with at most one overdamped half-cycle. The absence of cycles is presumably due to features in this 3-D model system that are absent in the 0-D models: powerful Height Mass-Balance Feedback producing strong ice-sheet expansion after initial growth, and hysteresis in ice-sheet response to climate that damps retreat due to moderate warming. With one of the weathering parameterizations, the models indicate a region of negative slope in the relation between CO2 level and global weathering consumption, occurring in the range 0.2 to 1.5x PAL (preindustrial atmospheric level). This contrasts with the monotonically increasing relation usually assumed. If confirmed, it would have serious consequences for the well-known CO2-weathering thermostat mechanism, at least for CO2 levels below 1.5x PAL.
keywords: Eocene–Oligocene Transition; Antarctic ice sheet; paleoclimate modeling; silicate weathering feedback; carbon cycles