On the effect of Southern Ocean eddies on ocean carbon storage and atmospheric pCO2

The mechanism behind the ~80-100 ppmv change in atmospheric pCO₂ over a glacial/inter-glacial cycle of the Earth system is yet to be fully realised. However, the close relationship between Antarctic temperature proxies and pCO₂ variations suggests that upwelling in the Southern Ocean may well play a crucial role. This upwelling is the net result of a wind induced Eulerian-mean upwelling and an eddy-induced downwelling, which together set the residual upwelling experienced by tracers in the Southern Ocean. The eddy-induced component of this residual involves a number of processes in a subtle balance with each other. However, here we concentrate on just the wind forcing element, which can potentially alter both the Eulerian-mean upwelling and the eddy-induced upwelling, by exciting the mesoscale eddy field. In particular, we examine the effect of increased and decreased wind forcing on the ocean circulation, and the resulting atmospheric pCO₂, in an idealised model configuration.

We use MITgcm in a narrow sector configuration to investigate the changes in circulation that occur at both coarse (climate model) resolutions and higher, eddy-permitting, resolutions. At coarse resolutions, the mesoscale eddy field is represented by the Gent & McWilliams parameterisation. However, at eddy-permitting resolutions, large geostrophic eddies are well represented by the model grid and we use the GM parameterisation as a sub-grid scale closure in order to maintain the adiabatic nature of the circulation. By coupling these physical circulations to MITgcm's simple biogeochemistry package, we are able to elucidate the effect that changes in the mesoscale eddy field, and/or its representation, have on atmospheric pCO₂. The narrow sector configuration allows us to perform a large number of experiments, over a wide range of parameter space and resolutions, whilst still reaching thermodynamic and biogeochemical equilibrium.

Recent observations have shown that the rate of uptake of anthropogenic CO₂ is not increasing as fast as expected from steady-state models, and modelling studies using non-eddy-resolving ocean models forced with observed winds have shown this is consistent with the effect of strengthening and poleward shifting winds on the ocean circulation. Our sector model configuration allows us to investigate the effects of changes in wind forcing on multiple timescales from interannual (eg the SAM) to decadal and century timescales. This can help us to understand the differences in response with resolved eddies compared to parameterised representations, and the effect on air-sea gas exchange and anthropogenic CO₂ uptake of the same.