Starting in 2012, the eastern subpolar North Atlantic experienced the strongest surface freshening in the past 120 years. It is yet unknown whether this salinity anomaly propagated downward into the water column and affected the properties of the boundary currents of the subpolar gyre, which could slow down the overturning. Here, we investigate the imprint of this salinity anomaly on the warm and saline Irminger Current (IC) in the decade thereafter. Using daily mooring data from the IC covering the period 2014–2022 combined with hydrographic sections across the adjacent basins from 1990, the evolving signal of the salinity anomaly over the water column and its imprint on the transport variability is studied. We find that due to the salinity anomaly, the northward freshwater transport of the IC increased by 10 mSv in summer 2016 compared to summer 2015. In 2018, the salinity anomaly covered the water column down to 1,500 m depth. Hydrographic sections across the basin showed that this recent freshening signal spread across the Irminger Sea. Overall, the freshwater transport of the IC increased by a factor of three between 2014–2015 and 2021–2022. The associated density decrease over the upper 1,500 m of the water column resulted in an increase in the northward transport of waters lighter than σ₀ = 27.55 kg m⁻³ from 1.7 to 4.2 Sv. This change in northward IC transport by density class may impact the characteristics of the overturning in the Northeastern Atlantic, its strength and the density at which it peaks.

From 1969 to 1971 convection in the Labrador Sea shut down, thus interrupting the formation of the intermediate/dense water masses. The shutdown has been attributed to the surface freshening induced by the Great Salinity Anomaly (GSA), a freshwater anomaly in the subpolar North Atlantic. The abrupt resumption of convection in 1972, in contrast, is attributed to the extreme atmospheric forcing of that winter. Here oceanic and atmospheric data collected in the Labrador Sea at Ocean Weather Station Bravo and a one-dimensional mixed layer model are used to examine the causes of the shutdown and resumption of convection in detail. These results highlight the tight coupling of the ocean and atmosphere in convection regions and the need to resolve both components to correctly represent convective processes in the ocean. They are also relevant to present-day conditions given the increased ice melt in the Arctic Ocean and from the Greenland Ice Sheet. The analysis herein shows that the shutdown was initiated by the GSA-induced freshening as well as the mild 1968/69 winter. After the shutdown had begun, however, the continuing lateral freshwater flux as well as two positive feedbacks [both associated with the sea surface temperature (SST) decrease due to lack of convective mixing with warmer subsurface water] further inhibited convection. First, the SST decrease reduced the heat flux to the atmosphere by reducing the air–sea temperature gradient. Second, it further reduced the surface buoyancy loss by reducing the thermal expansion coefficient of the surface water. In 1972 convection resumed because of both the extreme atmospheric forcing and advection of saltier waters into the convection region.

In the late 1960’s an anomalously fresh water mass, known as the Great Salinity Anomaly (GSA), reached the Labrador Sea. Coincident with this arrival, between 1968 and 1971 no intermediate to deep convection took place in this area. As a result, the Labrador Sea became increasingly more stratified, due to the accumulation of cold and fresh water at the surface and warm and salty Irminger Water beneath. Then, in the winter of 1972, deep convection resumed accompanied by the largest surface heat flux on record. This sequence of events has led to the widespread notion that convection stopped as a result of the increase in stratification due to the GSA and later resumed simply as a result of the extreme atmospheric fluxes in 1972. Here we examine both these assertions to investigate their actual causal relationships. In particular, we explore to what extent the shutdown of convection was due to oceanic processes alone, as there seems to be an important contribution from the sizable decrease in the atmospheric fluxes during this 1968-1971 period as well. Similarly, we investigate to what extent the large air-sea fluxes in 1972 were associated with the resumption of deep convection and the exposure of the stored warm, salty subsurface waters to the cold atmosphere. In other words, what really stopped deep convection and what made it come back? To address these questions we use a combination of oceanic hydrographic data collected by Ocean Weather Ship Bravo from 1964 to 1974 and of atmospheric data from reanalysis data products, in conjunction with a one-dimensional mixed-layer model to simulate the evolution of a water column in the central Labrador Sea over winter. Our results are relevant to our general understanding of which factors control the interannual variability of convection in the Labrador Sea (an important contributor to the Atlantic Meridional Overturning Circulation). Especially at a time when changes in both the subtropical and Arctic regions are contributing to large changes in stratification within the subpolar regions, the Great Salinity Anomaly is no longer just of historical interest.