By ventilating the deep ocean, deep convection in the Labrador Sea plays a crucial role in the climate system. Unfortunately, the mechanisms leading to the cessation of convection and, hence, the mechanisms by which a changing climate might affect deep convection remain unclear. In winter 2020, three autonomous underwater gliders sampled the convective region and both its spatial and temporal boundaries. Both boundaries are characterised by higher sub-daily mixed-layer depth variability sampled by the gliders than the convective region. At the convection boundaries, buoyant intrusions–including eddies and filaments–instead of atmospheric warming primarily trigger restratification by bringing buoyancy with a comparable contribution from either fresh or warm intrusions. At the edges of these intrusions, submesoscale instabilities, such as symmetric instabilities and mixed-layer baroclinic instabilities, seem to contribute to the decay of the intrusions. In winter, enhanced lateral buoyancy gradients are correlated with strong destabilising surface heat fluxes and along-front winds. Consequently, winter atmospheric conditions and buoyant intrusions participate in halting convection by triggering restratification while surface fluxes are still destratifying. This study reveals freshwater anomalies in a narrow area offshore of the Labrador Current and near the convective region; this area has received less attention than the more eddy-rich West Greenland Current, but is a potential source of freshwater in closer proximity to the region of deep convection. Freshwater fluxes from the Arctic and Greenland are expected to increase under a changing climate, and our findings suggest that they may play an active role in the restratification of deep convection.

No abstract

Abstract. Ocean gliders can provide high-spatial- and temporal-resolution data and target specific ocean regions at a low cost compared to ship-based measurements. An important gap, however, given the need for carbon measurements, is the lack of capable sensors for glider-based CO2 measurements. We need to develop robust methods to evaluate novel CO2 sensors for gliders. Here we present results from testing the performance of a novel CO2 optode sensor (Atamanchuk et al., 2014), deployed on a Slocum glider, in the Labrador Sea and on the Newfoundland Shelf. This paper (1) investigates the performance of the CO2 optode on two glider deployments, (2) demonstrates the utility of using the autonomous SeaCycler profiler mooring (Send et al., 2013; Atamanchuk et al., 2020) to improve in situ sensor data, and (3) presents data from moored and mobile platforms to resolve fine scales of temporal and spatial variability of O2 and pCO2 in the Labrador Sea. The Aanderaa CO2 optode is an early prototype sensor that has not undergone rigorous testing on a glider but is compact and uses little power. Our analysis shows that the sensor suffers from instability and slow response times (τ95>100 s), affected by different behavior when profiling through small (<3 ∘C) vs. large (>10 ∘C) changes in temperature over similar time intervals. We compare the glider and SeaCycler O2 and CO2 observations and estimate the glider data uncertainty as ± 6.14 and ± 44.01 µatm, respectively. From the Labrador Sea mission, we point to short timescales (<7 d) and distance (<15 km) scales as important drivers of change in this region.

&lt;p&gt;By ventilating the deep ocean, deep convection in the Labrador Sea plays a crucial role in the climate system. Unfortunately, the mechanisms leading to the cessation of convection and, hence, the mechanisms by which a changing climate might affect deep convection remain unclear. In winter 2020, three autonomous underwater gliders sampled the convective region and both its spatial and temporal boundaries. Both boundaries are characterised by higher sub-daily mixed-layer depth variability than the convective region. At the convection boundaries, buoyant intrusions--including eddies and filaments--primarily drive restratification by bringing freshwater, instead of warm warmer, and instead of atmospheric warming. At the edges of these intrusions, submesoscale instabilities, such as symmetric instabilities and mixed-layer baroclinic instabilities, seem to contribute to the decay of the intrusions. In winter, strong destabilising surface heat flux and along-front winds can enhance the lateral stratification, sustaining submesoscale instabilities. Consequently, winter atmospheric conditions and freshwater intrusions participate in halting convection by adding buoyant freshwater into the convective region through submesoscale flows. This study reveals freshwater anomalies in a narrow area offshore of the Labrador Current and near the convective region; this area has received less attention than the more eddy-rich West Greenland Current, but is a potential source of freshwater in closer proximity to the region of deep convection. Freshwater fluxes from the Arctic and Greenland are expected to increase under a changing climate, and our findings suggest that they may play an active role in the restratification of deep convection.&lt;/p&gt;