The hypothesis of this study was that there is a potential for an oscillating covariability between the freshwater content of the Arctic Ocean and the Subarctic North Atlantic, and the redistribution between their basins is governed by wind stress forcing associated with large-scale patterns of atmospheric variability. In order to test this hypothesis, numerical model simulations were performed using the Max Planck Institute Earth System Model (MPI-ESM) with the objectives to 1) analyze the link between Arctic and Subarctic North Atlantic freshwater anomalies, to 2) identify key patterns of atmospheric variability that govern these anomalies through wind forcing, and to 3) explain the physical mechanisms of coupling between freshwater and near-surface winds associated with these key patterns. The results showed that even though there is a stable sign of freshwater redistribution between the Arctic and the Subarctic North Atlantic on a multidecadal time scale, this sign is mostly obscured by large anomalies in the North Atlantic that are transported from the south. A comprehensive statistical analysis revealed that the main statistical modes of large-scale atmospheric variability do not represent those modes that are best connected to freshwater anomalies. Such modes were identified in this work by performing a redundancy analysis of atmospheric variability and freshwater content, separately for its liquid and solid components. The impact of wind stress forcing was demonstrated by further simulations. These used prescribed wind data based on observations, but unlike the otherwise identically set- up unconstrained fully coupled runs, they could reproduce the observed freshwater anomalies of the 1990s. This confirmed the key role of wind stress forcing. Additional experiments with prescribed idealized wind perturbations enabled the isolation of the effect of certain wind forcing patterns on freshwater variability.