Two-timescale carbon cycle response to an AMOC collapse

Atmospheric CO2 concentrations (pCO2) varied on millennial timescales in phase with Antarctic temperature during the last glacial period. A prevailing view has been that carbon release and uptake by the Southern Ocean dominated this millennial‐scale variability in pCO2. Here, using Earth System Mode...

Full description

Bibliographic Details
Published in:Paleoceanography and Paleoclimatology
Main Authors: Nielsen, SB, Jochum, M, Pedro, JB, Eden, C, Nuterman, R
Format: Article in Journal/Newspaper
Language:English
Published: Wiley-Blackwell Publishing Inc. 2019
Subjects:
Online Access:https://eprints.utas.edu.au/33257/
https://eprints.utas.edu.au/33257/1/137184%20-%20Two-timescale%20carbon%20cycle%20response%20to%20an%20AMOC%20collapse.pdf
Description
Summary:Atmospheric CO2 concentrations (pCO2) varied on millennial timescales in phase with Antarctic temperature during the last glacial period. A prevailing view has been that carbon release and uptake by the Southern Ocean dominated this millennial‐scale variability in pCO2. Here, using Earth System Model experiments with an improved parameterization of ocean vertical mixing, we find a major role for terrestrial and oceanic carbon releases in driving the pCO2 trend. In our simulations, a change in Northern Hemisphere insolation weakens the Atlantic Meridional Overturning Circulation (AMOC) leading to increasing pCO2 and Antarctic temperatures. The simulated rise in pCO2 is caused in equal parts by increased CO2 outgassing from the global ocean due to a reduced biological activity and changed ventilation rates, and terrestrial carbon release as a response to southward migration of the Intertropical Convergence Zone. The simulated terrestrial release of carbon could explain stadial declines in organic carbon reservoirs observed in recent ice core δ13C measurements. Our results show that parallel variations in Antarctic temperature and pCO2 do not necessitate that the Southern Ocean dominates carbon exchange; instead, changes in carbon flux from the global ocean and land carbon reservoirs can explain the observed pCO2 (and δ13C) changes.