Description: Changes in the Atlantic Meridional Overturning Circulation, which have the potential to drive societally-important climate impacts, have traditionally been linked to the strength of deep water formation in the subpolar North Atlantic. Yet there is neither clear observational evidence nor agreement among models about how changes in deep water formation influence overturning. Here, we use data from a trans-basin mooring array (OSNAP—Overturning in the Subpolar North Atlantic Program) to show that winter convection during 2014–2018 in the interior basin had minimal impact on density changes in the deep western boundary currents in the subpolar basins. Contrary to previous modeling studies, we find no discernable relationship between western boundary changes and subpolar overturning variability over the observational time scales. Our results require a reconsideration of the notion of deep western boundary changes representing overturning characteristics, with implications for constraining the source of overturning variability within and downstream of the subpolar region.

Description: We use a 30-year time series (1986–2016) of dichlorodifluoromethane (CFC-12) concentrations with a refined transit time distribution (TTD) method, to estimate the temporal variation of anthropogenic carbon (Cant) in the Central Labrador Sea. We determined that the saturation of CFC-12 and sulfur hexafluroide (SF6) in newly-formed Labrador Sea Water had departed significantly from 100% and varied systematically with time. Multiple linear regression of the time-varying saturation, with the tracer's atmospheric growth rate and the wintertime mixed layer depth as independent variables, allowed reconstruction of the saturation history of CFC-12 and SF6 in wintertime surface waters, which was implemented in the TTD method. Use of the time-varying saturation for CFC-12 gave Cant concentrations ∼7 μmol kg−1 larger than estimates obtained assuming a constant saturation of 100%. The resulting Cant column inventories were ∼20% larger and displayed lower interannual variability compared to conventional TTD-based estimates. The column inventory of Cant increased at an average rate of 1.8 mol m−2 y−1 over the 30-year period. However, the accumulation rate of Cant was higher than this average in the early 1990s and since 2013, whereas inventories remained almost unchanged between 2003 and 2012. The variation in the Cant accumulation rate is shown to be linked to temporal variability in the relative layer thickness of the annually ventilated Labrador Sea Water and the underlying Deep Intermediate Water. The non-steady Cant accumulation highlights the importance of sampling frequency, especially in regions of variable deep mixing and high carbon inventories, and potential misinterpretation of Cant dynamics