Large Scale Impacts of Subarctic Channel Flows and Deep Water Formation in Climate Models

Abstract

This thesis evaluates the impact of the channel flows in the Subarctic and dense water formed in the North Atlantic on large scale circulations and the climate. Here we focus on the processes related to the formation and transformation of two dense water sources: the dense water formed in the Nordic Seas and the Labrador Sea. Two fully-coupled global climate models (CM2G and CM2M) are utilized for analysis and experiments.

The climatology of the models is first analyzed with a focus on North Atlantic dense water and the Atlantic Meridional Overturning Circulation (AMOC). Due to their different ocean components, the two models exhibit differences in the AMOC and North Atlantic climate, which are caused by and related to the differences in dense water formation, including the Nordic Seas overflows and freshwater pathways into the Labrador Sea.

The impact of the Nordic Seas overflows representation is then studied. Two major differences related to the overflows are identified: (1) The Iceland-Scotland channels overflow is not included in CM2M. (2) The diapycnal diffusivity downstream of the overflows is much higher in CM2M. Perturbation experiments in CM2G are performed to evaluate the climatic impact of these two differences. It is shown that the inclusion of IS overflow is accompanied with a further eastward extended subpolar gyre, which leads to significant changes in surface climate. The vertical structure of the AMOC is found to be sensitive to the diapycnal mixing downstream of the overflows. In response to an enhanced diffusivity, the AMOC upper cell becomes shallower and Antarctic Bottom Water extends further to the north.

Next the relative roles of two Arctic freshwater export pathways on the Labrador Sea convection and the AMOC are studied. Perturbation experiments are performed to alter the relative freshwater transport via the two pathways in both CM2G and CM2M. It is found the Labrador Sea convection is more sensitive to the relatively indirect eastern route. The AMOC, on the other hand, does not necessarily follow the changes in the Labrador Sea convection, as the dense water formed in other regions tends to compensate for the change in Labrador Sea convection.