Significant multi-centennial climate variability with a clear peak at approximately 200 years is found in a pre-industrial control simulation conducted with the EC-Earth3 climate model. The oscillation mainly emerges from the North Atlantic and appears to be closely associated with the Atlantic Meridional Overturning Circulation (AMOC). By examining the salinity advection feedback, we find that the perturbation flow of mean subtropical-subpolar salinity gradients in the subpolar area governs as positive feedback to the AMOC anomaly. Meanwhile, the mean advection of salinity anomalies and the vertical mixing or convection acts as negative feedback to restrain the AMOC anomaly. In a warmer climate, although the AMOC becomes weaker, such low-frequency variability still exists, indicating the robustness of the salinity advection feedback mechanism.

Arctic sea ice has substantially changed over the past four decades, with a large decrease in sea-ice area and volume. The exact causes of these changes are not entirely known. In our study, we make use of the Swedish Meteorological and Hydrological Institute Large Ensemble (SMHI-LENS). This ensemble consists of 50 members realized with the EC-Earth3 global climate model and covers the period 1970-2100. We apply the Liang-Kleeman information flow method to analyze the cause-effect relationships between Arctic sea ice and its potential drivers. We show that recent and future changes in Arctic sea ice are mainly driven by air and sea-surface temperatures and ocean heat transport. Conversely, changes in Arctic sea ice also considerably impact temperature and ocean heat transport. Finally, we find a progressive decrease in the influence of sea-ice area and volume on air temperature and ocean heat transport through the twenty-first century.

Arctic sea ice has substantially changed over the past four decades, with a large decrease in sea-ice area and volume. The exact causes of these changes are not entirely known. In our study, we make use of the Swedish Meteorological and Hydrological Institute Large Ensemble (SMHI-LENS). This ensemble consists of 50 members realized with the EC-Earth3 global climate model and covers the period 1970-2100. We apply the Liang-Kleeman information flow method to analyze the cause-effect relationships between Arctic sea ice and its potential drivers. We show that recent and future changes in Arctic sea ice are mainly driven by air and sea-surface temperatures and ocean heat transport. Conversely, changes in Arctic sea ice also considerably impact temperature and ocean heat transport. Finally, we find a progressive decrease in the influence of sea-ice area and volume on air temperature and ocean heat transport through the twenty-first century.

The pandemic in 2020 caused an abrupt change in the emission of anthropogenic aerosols and their precursors. We provide the first estimate of the associated change in the aerosol radiative forcing at the top of the atmosphere and the surface. To this end, we perform new simulations with the contemporary Earth system model EC-Earth3 participating in CMIP6, and created new data on the anthropogenic aerosol optical properties and an associated effect on clouds for the implemented aerosol parameterization, MACv2-SP. Our results highlight the small impact of the pandemic on the global aerosol radiative forcing in 2020 compared to the baseline of the order of +0.04Wm-2, which is small compared to the natural year-to-year variability in the radiation budget. Natural variability also limits the ability to detect a meaningful regional difference in the anthropogenic aerosol radiative effects. We identify the best chances to find a significant change in radiation at the surface during cloud-free conditions for regions that were strongly polluted in the past years. The new post-pandemic recovery scenarios indicate a spread in the aerosol forcing of -0.68 to -0.38Wm-2 for 2050, which translates to a difference of +0.05 to -0.25Wm-2 compared to the baseline. This spread falls within the present-day uncertainty in aerosol radiative forcing and the CMIP6 spread in aerosol forcing at the end of the 21st century. We release the new MACv2-SP data for studies on the climate response to the pandemic and the recovery scenarios.