Eddy-topography interactions drive ocean ventilation across the Eocene–Oligocene transition: insights from a hierarchy of ocean model simulations

Résumé fourni par la source

Abstract The Eocene–Oligocene transition (EOT), approximately 34 million years ago, represents one of the most profound climate shifts in Earth’s history—a rapid cooling that marked the transition from the warm Eocene greenhouse to the Cenozoic icehouse. The mechanisms driving this transition remain debated: proposed causes include a decline in atmospheric CO 2 concentrations and tectonic opening of Southern Ocean gateways, namely Drake Passage and the Tasman Gateway, which altered ocean circulation and poleward heat transport. Here we use ocean model simulations spanning a wide range of spatial resolutions—from a coarse-resolution, eddy-parameterised configuration to high-resolution, eddy-resolving simulations—to investigate how the representation of ocean eddies affects the simulated circulation response to gateway opening. Our results show that the establishment of a proto Antarctic Circumpolar Current (ACC), even if much weaker than today’s current, led to the onset of coastal upwelling along the Antarctic continental slope in the Australian Antarctic Basin (AAB). This upwelling has previously been linked to the onset of biogenic blooms and associated carbon drawdown in the Southern Ocean. The simulations suggest that the proto-ACC followed the Antarctic continental slope in the AAB and that upwelling was driven by either bottom Ekman transport beneath this slope current or, as indicated by the highest-resolution simulations, eddy-topography interactions—a process in which mesoscale eddies interacting with rough and steeply sloping bottom topography drive a net onshore transport of dense waters. Critically, neither a distinct continental slope current nor the associated upwelling is reproduced in coarse-resolution models using existing eddy parameterisations, implying that the climatic impact of Southern Ocean gateway opening during the EOT has been substantially underestimated in previous modelling studies.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.