The sea surface salinity (SSS) variability in the North Atlantic is investigated using numerical model simulations for the last 50 years based on atmospheric forcing variability from the Comprehensive Atmosphere Ocean Data Set (COADS) and National Center for Environmental Prediction/National Center for Atmospheric Research (NCEP/NCAR) reanalysis. The focus here is the SSS variability in the subpolar region. The seasonality of the subpolar surface salinity variability is prominent with the maximum standard deviation (SD) occurring in the summer/fall period so that the amplitude of the summer SSS anomalies far exceed those of the wintertime. The interannual SSS variability in the subpolar gyre can be attributed mainly to excess ice melt, while the longer-term (decadal) variability is associated with meridional overturning variability. In these two hindcasts, the deep mixing (which drives overturning changes) is forced by heat flux variability so that weak subpolar heat loss and subsequent weak overturning are usually manifested in fresh surface conditions in the subpolar gyre within 2 years. This is because the role of deep convection is to mix down the net fresh water input received by the high latitudes. Hence, at the lack of widespread deep mixing, a signal of slowing down overturning, the fresh water input that would be expected based on climatology would accumulate at the surface.