GC117-0001
Causes and Impacts of the Inter-Model Spread in AMOC Change in Projections of Future Climate Change

Wednesday, 16 December 2020
Poster
Katinka Bellomo1, Jost von Hardenberg1,2, Michela Angeloni1,3, Virna Loana Meccia4, Susanna Corti4 and Mark Zelinka5, (1)CNR Institute of Atmospheric Sciences and Climate, Turin, Italy, (2)Polytechnic University of Turin, Department of Environment, Land and Infrastructure Engineering, Turin, Italy, (3)University of Bologna, Department of Physics and Astronomy, Bologna, Italy, (4)CNR Institute of Atmospheric Sciences and Climate, Bologna, Italy, (5)Lawrence Livermore National Laboratory, Livermore, CA, United States
Abstract:
In response to increasing concentrations of anthropogenic greenhouse gases, climate models predict an increase in global mean surface temperature and a decline of the Atlantic Meridional Overturning Circulation (AMOC) in the ocean. Here we use 30 idealized abrupt-4xCO2 climate model simulations from the CMIP5 and CMIP6 archives to demonstrate that the inter-model range in the decline rates of the AMOC drives different responses in transient climate change, especially over the North Atlantic and Arctic regions. In models with larger AMOC declines, the North Atlantic surface temperature increases less than in the models with smaller AMOC declines. This difference can be regionally larger than 10K. In addition, in models in which the AMOC decline is larger, there is a shift of the ITCZ to the southern hemisphere, which is not seen in models in which the AMOC decline is smaller. While we find that the amplitude of the AMOC decline is related to the inter-model range in the simulation of the mean climate, there are many different variables that contribute. Hence, we use cluster analysis to group models based on their simulations of multiple variables in the mean climate, revealing the characteristics that lead to a smaller rather than a bigger AMOC decline in projections of future climate change. Our results further indicate that by constraining biases in the mean climate using observations, we could significantly reduce uncertainty in projections of future climate change in the North Atlantic.