Glacial-interglacial response of the global overturning circulation to southern forcing; carbon cycle applications

Abstract

Over the past one million years, atmospheric CO2 and Antarctic temperatures covaried over several glacial cycles, with the meridional overturning circulation (MOC) as forced by the climate over the Southern Ocean implicated as a causal factor. This thesis aims to improve our understanding of the mechanisms responsible for glacial-interglacial variations in the MOC and atmospheric CO2, and also the pathways and mechanisms driving the present-day, glacial and possible future state circulations. Both a semi-analytical ocean model and an idealised general circulation model (GCM) are used. An idealised two-basin and single-basin GCM connected by a southern circumpolar channel shows the MOC strength and structure is dependent on the Southern Ocean buoyancy and zonal wind forcing, and the ocean’s vertical diffusivity. The sensitivity of the MOC to wind and diffusivity perturbations is dependent on the buoyancy forcing, due to the MOC pathways varying with buoyancy forcing. The North Atlantic Deep Water (NADW) cell in the Atlantic basin is driven primarily by the Southern Ocean winds in a glacial state but by Pacific diffusive-driven upwelling in a future ‘warm’ state. The single-basin model is shown to have a number of drawbacks, and is unable to simulate a realistic NADW cell strength under realistic wind forcing and vertical diffusivity, regardless of the buoyancy forcing imposed. In contrast, the two-basin model simulates a realistic NADW cell, with a connection between the NADW cell in the Atlantic basin and the Pacific Deep Water (PDW) cell in the Pacific basin under present-day buoyancy forcings. A transition of the MOC to a glacial state with a weaker, shoaled NADW cell isolated from the Pacific basin is obtained in response to enhanced Southern Ocean sea-ice formation and thus a change in buoyancy forcing. However, reduced Southern Ocean wind forcing leads to only a slight weakening and no shoaling of the NADW cell and thus a glacial state MOC is not obtained. Changes in the buoyancy forcing between a future warm state and a cool, glacial state lead to a reduction in atmospheric CO2 of ~30 ppm in the two-basin model when temperature solubility effects are ignored. This is smaller than the ~100 ppm glacial-interglacial change in CO2, although the model is likely to be highly dependent on the biological parameters, and the effect of sea-ice on biological production and air-sea CO2 exchange. A semi-analytical ocean model with a coupled energy-balance model is also used to investigate the glacial cycles. The important role played by climate feedbacks in the MOC transition is apparent, with changes in Southern Ocean sea-ice formation playing a potentially critical role.