Key points are not available for this paper at this time.
This study aims to analyze the effect of increasing atmospheric CO2 concentrations on the Atlantic Meridional Overturning Circulation (AMOC) and its dependence on convection in the Labrador (LAB) and Greenland (GIN) Seas. We have used EC-Earth3-HR, the high-resolution version of the global coupled climate model EC-Earth3 in this study. EC-Earth3-HR has a resolution of about 0.25 degrees in the ocean and 40 km in the atmosphere. In contrast to the HighResMIP-protocol, EC-Earth3-HR has undergone a tuning process and a multi-centennial spin-up has been performed. The set of experiments analyzed here consists of a pre-industrial control simulation (piControl), a one percent per year increase in CO2 experiment (1pctCO2) branching from year 250 of our piControl simulation, and two experiments with fixed CO2 concentrations (400.9 ppm and 551.5 ppm) branch off from two points corresponding to global temperature anomalies of around 1C and 2C in the 1pctCO2 experiment. Here we have defined deep convection as the mean mixed volume in March, with deep convection equal to zero when the mixed-layer is shallower than a critical depth. Our preliminary results suggest that as the climate warms, the North Atlantic waters become warmer and fresher, promoting the weakening of the North Atlantic deep convection and a subsequent reduction in AMOC strength (up to 20% reduction). The simulated overturning circulation weakening seems to be dominated by changes in LAB deep convection with GIN convection contributing less. Circulation changes in the pre-industrial and the different CO2 concentration experiments are dominated by a strong decadal variability. Compared to the standard resolution EC-Earth3-version, the use of a high resolution leads to deeper ocean mixing in LAB and GIN. More analysis has to be done on the way to clarify to what extent increased resolution affects our results in comparison with previous studies.
Labastida et al. (Mon,) studied this question.