Investigating the impact of physical processes in the Southern Ocean on the carbon cycle during past, present and future climates
Final Report Abstract
A significant influence of changes in the westerly winds over the Southern Ocean was proposed as a mechanism to explain a large portion of the glacial atmospheric pCO2 draw down (Toggweiler et al., 2006). However, additional modelling studies with Earth System Models of Intermediate Complexity do not confirm the size and sometimes even the sign of the impact of southern hemispheric winds on the glacial pCO2 as suggested by Toggweiler. We here add to this discussion and explore the potential contribution of changes in the latitudinal position of the winds on Southern Ocean physics and the carbon cycle by using a state-of-the-art ocean general circulation model (MITgcm) in a spatial resolution increasing in the Southern Ocean. We discuss how the change in carbon cycling is related to the upwelling strength and pattern in the Southern Ocean and how they depend on the changing wind fields and/or the sea ice coverage. While the previous studies explored the impact of the westerlies starting from present day or preindustrial background conditions, we here perform additionally simulations for background climate of the Last Glacial Maximum (LGM, about 20,000 year before present). Ocean surface conditions are taken from output of the COSMOS Earth System model for the LGM run. Both a northwards and southwards shift of the westerly wind belt by 10° is investigated. Our results show, that the background climate has a significant influence on the simulations. A northwards shift of the Southern Ocean wind belt leads in both cases (LGM and present day) to a reduction of the Agulhas leakage and thus to a fresher Atlantic sea surface and consequently to reduced Atlantic meridional overturning circulation. A southwards shift in the wind belt leads during LGM background climate to an intensification of the ACC. Patterns how temperature and biogeochemistry (DIC) changes over depth are nearly opposite to those of d’Orgeville et al. (2010). Atmospheric CO2 rises by 5 ppmv for both cases, the northwards or southwards wind shift. In conclusion, we propose that the investigation of changes in the westerly wind belt in the Southern Ocean for present day is not comparable with LGM and simulation studies without LGM background climate are of limited use.
