The role of ocean circulation and sea ice in abrupt climate change

Abstract

Abrupt changes in Earth's climate have occurred repeatedly throughout the geological record. Evidence from paleoclimate data has revealed that climate changed most dramatically during the last glacial period, associated with the Dansgaard-Oeschger (D-O) events. These are characterized by large and rapid fluctuations in North Atlantic climate, with regional warming of up to 15 degrees C over Greenland, within a few decades. The main hypotheses for these abrupt climate changes in the past, centers around changes in the Atlantic Meridional Overturning Circulation (AMOC) and its influence on poleward ocean heat transport. Recently, the role of sea ice has also been recognized as a critical player for the D-O events; linking the abrupt changes in Greenland temperature to a retreat of Northern Hemisphere sea ice, driven by internal variability of the coupled ice-ocean and atmosphere system. However, the mechanism for triggering rapid changes in sea ice, and how it is linked to ocean circulation changes, remains elusive. This thesis focuses on the interaction between ocean circulation, sea ice and high-latitude climate in the context of abrupt climate changes in the past. The main goal is to improve our understanding of how internal dynamics of the coupled climate system can lead to rapid and unforced changes in climate. The first part of this thesis explores the mechanism behind abrupt changes in sea ice by studying the formation of open-ocean polynyas in the Southern Ocean. The second and third part, focuses on the dynamics of large-scale ocean circulation and its sensitivity to ocean bathymetry and the distribution of diapycnal mixing in the ocean interior. We find that open-ocean polynyas in the Southern Ocean provide a mechanism to trigger abrupt sea ice retreat, similar to that seen during the last glacial period. These events drive increased bottom water formation, thereby impacting the large-scale ocean circulation. The formation of the polynya is preconditioned by a gradual build-up of subsurface heat and salt beneath the ice cover. This destabilizes the water column, triggering enhanced vertical mixing and causing the polynya to open. Our findings suggest that open-ocean polynyas, driven by internal ocean-sea ice dynamics, may play a fundamental role in abrupt climate changes such as D-O events. It is demonstrated that ocean basin geometry has a major impact on ocean circulation. In particular, the presence of the Greenland-Scotland Ridge affects the location of deep water formation and plays a fundamental role in shaping the AMOC and high-latitude climate. Interestingly, the strength of the AMOC at 26N is largely decoupled from deep water formation north of the ridge, and the AMOC plays a relatively small role in transporting heat northward across the Greenland-Scotland Ridge. This calls into question the role of the AMOC as the main driver of past abrupt changes in high latitude climate. Finally, the distribution and magnitude of vertical mixing in the ocean is found to play a central role in the stability of the coupled climate system and for the existence of the D-O events. Unforced and self-sustained «DO-like» oscillations can occur when thermocline mixing is low and the AMOC is reduced, allowing heat to accumulate below the sea ice, thereby preconditioning the system for an abrupt change. In addition, we find that changes in the abyssal mixing do not have a large impact on AMOC strength and surface climate. In summary, the results presented in this thesis confirm that ocean circulation has played a persistent and central role in abrupt climate change in the past, but emphasizes that variations in AMOC strength might not be the main trigger. In addition, the thesis highlights changes in sea ice as a necessary condition to drive large and rapid changes in high latitude climate. Such changes may occur in response to unforced and self-sustained oscillations of the coupled atmosphere-ocean-sea ice system, demonstrating that abrupt climate change can occur without being subject to large external forcing. This has important implications for predicting future abrupt changes in climate as a response to anthropogenic forcing, noting that the dynamics of abrupt changes as seen in the past can also operate in a warming climate.