Large-scale control of the retroflection of the Labrador Current

The study investigates what controls the retroflection of the Labrador Current at the Grand Banks of Newfoundland and how this process redistributes cold, fresh, oxygen-rich waters between the subpolar North Atlantic and the eastern American continental shelf/Slope Sea. Recent decades have seen warming and deoxygenation on the northeastern American shelf and a strong freshening of the subpolar North Atlantic, changes linked to variability in the export pathways of Labrador Current Water. The Labrador Current, comprising inshore and offshore branches sourced from the West Greenland Current and Arctic outflow, splits near the Grand Banks—either retroflecting eastward to join the North Atlantic Current or continuing westward along the shelf. Despite its importance, the dynamics governing the magnitude and variability of this retroflection remain poorly understood. The authors propose a new Lagrangian retroflection index to quantify variability over the past ~25 years and to relate it mechanistically to large-scale drivers such as winds, subpolar gyre adjustment, Labrador Current strength, and Gulf Stream position, and to evaluate resulting temperature, salinity, and oxygen anomalies in export regions.

Prior work proposed several, sometimes conflicting, controls on Labrador Current retroflection: local wind patterns over the Labrador Shelf; the strength of the Labrador Current itself; links to the North Atlantic Oscillation (reported as both strong-NAO and weak-NAO associations in different studies); possible connections to Atlantic Meridional Overturning Circulation strength; and a relationship with a northward shift of the Gulf Stream/North Atlantic Current. Mesoscale interactions at the Tail of the Grand Banks—eddies and meanders from the Gulf Stream/NAC—have been suggested to divert Labrador Current waters offshore or impede their westward inflow toward the Scotian Shelf. Seasonal stratification near the Grand Banks has also been implicated in modulating freshwater export from the shelf. This study reconciles these perspectives by framing retroflection variability within a basin-scale adjustment of the subpolar gyre that links winds, Labrador Current strength, and Gulf Stream position.

The analysis is based on Lagrangian particle tracking using daily velocity fields from the global 1/12° ocean physical reanalysis GLORYS12V1 (1993–2018), run with NEMO and LIM2 and forced by ERA-Interim. GLORYS12V1 assimilates satellite SST, altimetry, in situ T/S profiles, and sea-ice concentrations, and reproduces key North Atlantic circulation features. To isolate Labrador Current waters from the Deep Western Boundary Current, only waters with practical salinity Sp < 34.8 were considered. Virtual passive particles were tracked using the OceanParcels tool with daily horizontal velocities and reconstructed vertical velocities consistent with non-divergence and sea surface height changes. Seeding: particles were released weekly from 01-01-1993 to 01-01-2015 along a line from (53.0°N, 56.7°W) to (54.3°N, 52.0°W), spaced every 1/12° horizontally and every 10 m vertically (966 particles per seeding). Particles were integrated with 10-minute time steps for 3 years per release. A retroflection index (1993–2015) was computed by daily counting particles crossing two hydrographic sections (on the Labrador Shelf and on the Scotian Shelf) and taking their difference; the index was detrended to remove a significant positive trend, normalized to [-1, 1], and smoothed (12-month running mean) for composite analyses. Best-lag relationships to upstream forcing were assessed, with an approximately 8-month advection lag between the Labrador Shelf and the Grand Banks used in comparisons. Observational validation used trajectories of 64 Argo floats (2001–2019), 50 RAFOS/SOFAR floats (2003–2007), and 79 surface drifters (2000–2018) selected to cross specified sections near the Grand Banks. Eulerian diagnostics (volume transports, density gradients, barotropic streamfunction over top 1000 m, SSH) were derived from GLORYS outputs. Large-scale atmospheric forcing analyses used ERA-Interim wind stress curl and sea level pressure, and indices for AO/NAO; AMOC transport at 26°N (from CMS) was compared. Eddy/meander influences near the tip of the Grand Banks were assessed using an Okubo–Weiss-based eddy detection algorithm applied to SSH and flow at ~185 m (threshold OW = -0.35; eddies smaller than 190 pixels excluded), with counts compared between strong/weak retroflection composites and qualitative trajectory-eddy interactions examined. Additional backtracking experiments from the Scotian Shelf/Slope Sea assessed source waters, confirming <20% origin from the Labrador Current in that region.

- The Labrador Current export exhibits a seesaw between a westward branch feeding the Slope Sea/eastern American shelf (~25% of transport downstream of the Grand Banks over 1993–2015) and an eastward retroflected branch joining the NAC (~60%). Most retroflection occurs between Flemish Cap and the tip of the Grand Banks and at the tip itself (approximately 25% and 30% particle loss from the shelf at these locations, respectively). - A Lagrangian retroflection index (1993–2015) tracks temperature and salinity anomalies in both export regions (correlation coefficients > 0.55, p < 0.001), confirming the seesaw behavior: strong retroflection is associated with positive salinity and temperature anomalies in the Slope Sea/Scotian Shelf and negative salinity anomalies across the subpolar North Atlantic, with freshwater spreading eastward with the NAC. - Quantitatively, a unit increase in the index is associated with a 0.10 decrease in salinity in the subpolar North Atlantic and a 0.05 increase in the Slope Sea/near Scotian Shelf. - The retroflection index shows strong interannual-to-multiannual variability (standard deviation ~22%) and a significant positive trend of +2.4% per decade. Weak periods: 1996–1998, 2007, 2008–2009; strong periods: 1994–1996, 1999, 2002, 2011–2014. The strong 2011–2014 period aligns with enhanced Labrador-origin waters reaching the eastern North Atlantic and the 2012–2016 subpolar freshening; weak 1996–1999 coincides with high subpolar salinities. - SSH anomalies near the Grand Banks/Scotian Shelf are positively associated with strong retroflection, indicating a northward shift of the Gulf Stream/NAC. A change point in 2008 marks a pronounced northward shift of the Gulf Stream concurrent with a shift toward stronger retroflection (2009–2015). However, retroflection also occurred in 1994–1996 when the Gulf Stream was further south, indicating GS proximity is not necessary. - Strong retroflection generally coincides with a stronger Labrador Current (upstream). Correlation between Labrador Shelf transport and the retroflection index is positive (r ≈ 0.42 for 1999–2016, p < 0.001); downstream transport on the Scotian Shelf correlates negatively, consistent with eastward diversion reducing westward supply. - Wind stress curl anomalies modulate retroflection: negative anomalies over the Labrador Shelf/Grand Banks (stronger, north-shifted westerlies; poleward shift of the zero wind-stress curl line) are associated with strong retroflection, contraction of the western subpolar gyre, and LC acceleration. Positive anomalies and a southward shift of zero curl link the Labrador Sea and Slope Sea, reducing offshore push and favoring weak retroflection. The 1996–1998 weak event occurred despite a strong LC and low SSH, likely due to such wind pattern connectivity. - Atmospheric pressure patterns during strong retroflection exhibit an enhanced meridional gradient resembling a positive AO-like configuration (weak negative contemporaneous correlations with AO and NAO: r ≈ -0.33 and -0.26; no lagged correlation with AMOC strength at 26°N). - Mesoscale features near the tip of the Grand Banks: anti-cyclonic eddies are ~15% more numerous during strong retroflection and ~15% fewer during weak events, consistent with a northward GS shift. However, particle trajectories show retroflection can occur without local anti-cyclones and westward motion can occur despite their presence; retroflected waters preferentially follow cyclonic meanders/eddies of the Labrador Current. - Best-lag correlations suggest an ~8-month advective timescale between upstream forcing on the Labrador Shelf and retroflection expression at the Grand Banks.

The results demonstrate that variability in the Labrador Current retroflection is primarily governed by large-scale adjustments of the subpolar gyre that jointly affect three linked drivers: westerly winds (and wind stress curl), the strength of the Labrador Current, and the position of the Gulf Stream/North Atlantic Current. Strong retroflection arises when an enhanced meridional atmospheric pressure gradient shifts and intensifies westerlies over the western basin, pushing the Labrador Current offshore near the Grand Banks, contracting the western subpolar gyre to accelerate the LC, and shifting the Gulf Stream northward. While local Gulf Stream eddies and meanders can block or steer flows, they are not necessary conditions for retroflection; instead, retroflected waters largely follow cyclonic meanders of the Labrador Current itself. Since 2008, the pronounced northward displacement of the Gulf Stream has amplified retroflection, reinforcing warm, salty, low-oxygen incursions into the Slope Sea/Scotian Shelf and enhancing freshwater export into the subpolar North Atlantic. These dynamics explain observed hydrographic anomalies in both export regions and connect to potential impacts on stratification, convection, and large-scale circulation.

This study introduces a Lagrangian retroflection index that robustly quantifies multi-decadal variability in the Labrador Current’s eastward retroflection and links it to large-scale atmospheric and oceanic drivers. The key contribution is a unified, mechanistic framework wherein subpolar gyre adjustment—mediated by wind stress curl anomalies—modulates Labrador Current strength and Gulf Stream position, thereby controlling retroflection. Strong retroflection periods correspond to increased SSH near the Grand Banks, stronger LC, poleward-shifted westerlies, and enhanced cyclonic meanders guiding eastward diversion, with an especially prominent role for the northward-shifted Gulf Stream since 2008. The index connects directly to temperature, salinity, and oxygen anomalies in both export regions, enabling monitoring and prediction of ecosystem-relevant changes. Future work should more closely examine anomalous periods (e.g., 1996–1998) to isolate causal mechanisms, refine representation of mesoscale processes, and integrate sustained observations (winds, moorings for LC strength, SSH) into an operational framework for predicting water mass exports and their ecological impacts.

- The analysis relies on a specific ocean reanalysis (GLORYS12V1) that, while skillful, exhibits known biases (e.g., slight overestimation of western boundary current intensity; underestimation of Labrador shelf-break jet velocities) and excludes tides; results focus on variability rather than absolute magnitudes to mitigate bias impacts. - The retroflection index depends on particle tracking choices (seeding line, depth distribution, release frequency, and tracking duration) and on the definition using two hydrographic sections; alternative choices could alter magnitudes though sensitivity tests indicate robustness. - Eddy detection employs thresholds and resolution constraints (Okubo–Weiss at ~185 m; exclusion of small eddies), potentially missing smaller-scale features and complicating quantitative attribution of mesoscale influences. - The study period for the index ends in 2015; changes after 2015 (including ongoing Gulf Stream shifts or atmospheric patterns) are not captured. - A full causal attribution of the 1996–1998 weak retroflection period is unresolved and identified as requiring more detailed analysis. - Separation of Labrador Current waters from the DWBC using Sp < 34.8 may exclude some relevant variability near the shelf-break and slope.