Mechanisms for late 20th and early 21st century decadal AMOC variability

Recent studies using data from the OSNAP observational campaign and from numerical ocean models suggest that the Iceland Basin and the Irminger Sea may be more significant for formation of upper North Atlantic Deep Water than the Labrador Sea. Here, we present a set of hindcast integrations of a global 1/4° NEMO simulation from 1958 until nearly the present day, forced with three standard forcing data sets. We use the surface-forced stream function, estimated from surface buoyancy fluxes, along with the overturning stream function, similarly defined in potential density space, to investigate the causal link between surface forcing and decadal variability in the strength of the Atlantic meridional overturning circulation (AMOC). We use the stream functions to demonstrate that watermasses in the simulations are transformed to higher densities as they propagate around the subpolar gyre from their formation locations in the north-east Atlantic and the Irminger Sea, consistent with the picture emerging from observations. The surface heat loss from the Irminger Sea is confirmed to be the dominant mechanism for decadal AMOC variability, with the heat loss anomaly from the Labrador Sea having about half the magnitude. A scalar metric based on the surface-forced stream function, accumulated in time, is found to be a good predictor of changes in the overturning strength. The AMOC variability is shown to be related to that of the North Atlantic Oscillation (NAO), primarily through the surface heat flux, itself dominated by the air-sea temperature difference, but also with some local feedback from the SST to the surface fluxes.