Contribution of enhanced Antarctic Bottom Water formation to Antarctic warm events and millennial-scale atmospheric CO2 increase

Abstract:

During Marine Isotope Stage 3, the Atlantic Meridional Overturning Circulation (AMOC) weakened significantly on a millennial time-scale leading to Dansgaard-Oeschger (DO) and Heinrich stadials. Ice core records reveal that each Northern Hemisphere stadial is associated with a warming over Antarctica, so-called Antarctic Isotope Maximum (AIM), and that atmospheric CO₂ varies in phase with Antarctic temperature.

Here we perform transient simulations spanning the period 50-34 ka B.P. with two Earth System Models (LOVECLIM and the UVic ESCM) to understand the link between changes in the AMOC, changes in high latitude Southern Hemispheric climate and evolution of atmospheric CO₂. Given the latest Antarctic ice core chronology, we find that part of the atmospheric CO₂ increase occurring during AIM12 (DO12, ~48 ka B.P.) and at the end of AIM8 (DO8, 38 ka B.P.) can be attributed to the AMOC resumption. In contrast, the atmospheric CO₂ increase observed at the beginning of AIM8 (~39.6 ka B.P.) occurs during a period of weak AMOC and can instead be explained by enhanced Antarctic Bottom Water production. Enhanced Antarctic Bottom Water formation is shown to effectively ventilate the deep Pacific carbon and thus lead to CO₂ outgassing into the atmosphere.

In addition, changes in the AMOC alone are not sufficient to explain the largest Antarctic Isotope Maxima (namely AIM12 and AIM8). Stronger formation of Antarctic Bottom Water during AIM12 and AIM8 enhances the southern high latitude warming and leads to a better agreement with high southern latitude paleoproxy records. The robustness of this southern warming response is tested using an eddy-permitting coupled ocean sea-ice model. We show that stronger Antarctic Bottom Water formation contributes to Southern Ocean surface warming by increasing the Southern Ocean meridional heat transport. Finally, our simulations also suggest that the Antarctic cooling should be in phase, or lag by a maximum of ~200 years, the North Atlantic warming.