The role of subpolar deep water formation and Nordic Seas overflows in simulated multidecadal variability of the Atlantic meridional overturning circulation

International audienceWe investigate the respective role of variations insubpolar deep water formation and Nordic Seas overflows forthe decadal to multidecadal variability of the Atlantic merid-ional overturning circulation (AMOC). This is partly doneby analysing long (order of 1000 years) control simulationswith five coupled climate models. For all models, the max-imum influence of variations in subpolar deep water forma-tion is found at about 45◦N, while the maximum influenceof variations in Nordic Seas overflows is rather found at 55to 60◦N. Regarding the two overflow branches, the influenceof variations in the Denmark Strait overflow is, for all mod-els, substantially larger than that of variations in the over-flow across the Iceland–Scotland Ridge. The latter might,however, be underestimated, as the models in general do notrealistically simulate the flow path of the Iceland–Scotlandoverflow water south of the Iceland–Scotland Ridge. The in-fluence of variations in subpolar deep water formation is,on multimodel average, larger than that of variations in theDenmark Strait overflow. This is true both at 45◦N, wherethe maximum standard deviation of decadal to multidecadalAMOC variability is located for all but one model, and atthe more classical latitude of 30◦N. At 30◦N, variations in subpolar deep water formation and Denmark Strait overflowexplain, on multimodel average, about half and one-third re-spectively of the decadal to multidecadal AMOC variance.Apart from analysing multimodel control simulations, wehave performed sensitivity experiments with one of the mod-els, in which we suppress the variability of either subpolardeep water formation or Nordic Seas overflows. The sensitiv-ity experiments indicate that variations in subpolar deep wa-ter formation and Nordic Seas overflows are not completelyindependent. We further conclude from these experimentsthat the decadal to multidecadal AMOC variability north ofabout 50◦N is mainly related to variations in Nordic Seasoverflows. At 45◦N and south of this latitude, variations inboth subpolar deep water formation and Nordic Seas over-flows contribute to the AMOC variability, with neither of theprocesses being very dominant compared to the other