Early interglacial legacy of deglacial climate instability

• Published in: Paleoceanography and Paleoclimatology
• Main Authors: Barker, Stephen, Knorr, Gregor, Conn, Stephen, Lordsmith, Sian, Newman, Dhobasheni, Thornalley, David
• Format: Article in Journal/Newspaper
• Language: English
• Published: American Geophysical Union 2019
• Subjects: North Atlantic
• Online Access: https://orca.cardiff.ac.uk/id/eprint/125167/; https://doi.org/10.1029/2019PA003661; https://orca.cardiff.ac.uk/id/eprint/125167/1/855081_1_merged_1564606932.pdf

Throughout the last glacial cycle millennial timescale variations in atmospheric CO2 occurred in parallel with perturbations in deep ocean circulation, which were themselves reflected by observable changes in surface conditions across the North Atlantic region. Here we use continuous proxy records to argue that an equivalent relationship has held throughout the last 800 kyr, that is, since before the first occurrence of Heinrich events (strictly defined). Our results highlight the importance of internal climate dynamics in amplifying external (insolation) forcing on the climate system to produce the large amplitude of glacial terminations (deglaciations) during the middle to late Pleistocene. We show that terminations are characterized by an interval of intense ice rafting followed by a subsequent and abrupt shift to anomalously warm surface conditions (with respect to the more gradually evolving background state), which we interpret to reflect an abrupt recovery of deep ocean circulation in the Atlantic. According to our synthesis, this is followed by a period of enhanced (or at least anomalous) overturning lasting thousands of years until equilibrium interglacial conditions are attained and during which atmospheric CO2 is likely to decrease. Our results therefore suggest that deglacial oscillations in ocean circulation can have a lasting influence on early interglacial climate and highlight the transient nature of atmospheric CO2 overshoots associated with the onset of some previous interglacials. Accordingly, we suggest that these intervals should be considered as a part of the deglacial process. This has implications for studies concerned with the evolution of atmospheric CO2 during interglacial periods including the Holocene.