<p>The Atlantic Meridional Overturning Circulation (AMOC) has been, and will continue to be, a key factor in the modulation of climate change both locally and globally. Reliable simulations of its decadal to century-timescale evolution are key to providing skilful predictions of future regional climate, and to understanding the likelihood of a potential AMOC collapse. However, there remains considerable uncertainty even in past AMOC evolution. Here, we show that the multi-model mean AMOC strengthened by approximately 10% to 1985 in new historical simulations for the 6th Coupled Model Inter-comparison Project (CMIP6), contrary to results obtained from CMIP5. The simulated strengthening is due to a stronger anthropogenic aerosol forcing, in particular due to aerosol-cloud interactions. However, comparison with an observed sea surface temperature fingerprint of AMOC evolution during 1850-1985, and the shortwave forcing during 1985-2014, suggest that anthropogenic forcing and the subsequent AMOC response may be overestimated in some CMIP6 models.</p>

Previous work has shown that anthropogenic aerosol (AA) forcing drives a strengthening in the Atlantic Meridional Overturning Circulation (AMOC) in CMIP6 historical simulations over 1850–1985, but the mechanisms have not been fully understood. Across CMIP6 models, it is shown that there is a strong correlation between surface heat loss over the subpolar North Atlantic (SPNA) and the forced strengthening of the AMOC. Despite the link to AA forcing, the AMOC response is not strongly related to the contribution of anomalous downwelling surface shortwave radiation to SPNA heat loss. Rather, the spread in AMOC response is primarily due to the spread in turbulent heat loss. We hypothesize that turbulent heat loss is larger in models with strong AA forcing because the air advected over the ocean is colder and drier, in turn because of greater AA forced cooling over the continents upwind, especially North America. The strengthening of the AMOC also feeds back on itself positively in two distinct ways: by raising the sea surface temperature and hence further increasing turbulent heat loss in the SPNA, and by increasing the sea surface density across the SPNA due to increased northward transport of saline water. A comparison of key indices suggests that the AMOC response in models with strong AA forcing is not likely to be consistent with observations.

<p>The observational network around the North Atlantic has improved significantly over the last few decades with the advent of Argo and satellite observations, and the more recent efforts to monitor the Atlantic Meridional Overturning Circulation (AMOC) using arrays such as RAPID and OSNAP. These have shown decadal timescale changes across the North Atlantic including in heat content, heat transport and the circulation. </p><p>However there are still significant gaps in the observational coverage, and significant uncertainties around some observational products. Ocean reanalyses integrate the observations with a dynamically consistent ocean model and are potentially tools that can be used to understand the observed changes. However the suitability of the reanalyses for the task must also be assessed.<br>We use an ensemble of global ocean reanalyses in comparison with observations in order to examine the mean state and interannual-decadal variability of the North Atlantic ocean since 1993. We assess how well the reanalyses are able to capture different processes and whether any understanding can be inferred. In particular we look at ocean heat content, transports, the AMOC and gyre strengths, water masses and convection. </p><p> </p>

Atlantic multidecadal variability (AMV) has been linked to the observed slowdown of global warming over 1998–2012 through its impact on the tropical Pacific. Given the global importance of tropical Pacific variability, better understanding this Atlantic–Pacific teleconnection is key for improving climate predictions, but the robustness and strength of this link are uncertain. Analyzing a multi-model set of sensitivity experiments, we find that models differ by a factor of 10 in simulating the amplitude of the Equatorial Pacific cooling response to observed AMV warming. The inter-model spread is mainly driven by different amounts of moist static energy injection from the tropical Atlantic surface into the upper troposphere. We reduce this inter-model uncertainty by analytically correcting models for their mean precipitation biases and we quantify that, following an observed 0.26 °C AMV warming, the equatorial Pacific cools by 0.11 °C with an inter-model standard deviation of 0.03 °C.