The interannual variability of the surface eddy kinetic energy in the Labrador Sea

The variability of the surface eddy kinetic energy (EKE) in the Labrador Sea is investigated with a suite of numerical integrations using a regional ocean model. Simulations are performed over the period 1980–2001 and are compared to satellite observations over the last 9 years. The surface EKE pattern in the basin is dominated by a region along the West coast of Greenland where eddies, mainly anticyclonic, are formed by instability of the main currents flowing over the continental slope, consistent with previous idealized results. Here the interannual changes are linked to the shear of the incoming boundary current system imposed as boundary condition to the model domain. The highly variable strength of the East Greenland current at the northeast boundary, derived from the Simple Ocean Data Assimilation (SODA) reanalysis, strongly influences the vortex formation.In the center of the Labrador Sea, where deep convection occurs, a statistically significant portion of the modeled interannual surface EKE variability is correlated with the local atmospheric forcing, and both heat and wind fluxes play an important role and can be adopted as predictors at a lag of 2–3 months. The Arctic Oscillation index can also be used as a remote indicator of the atmospheric fluxes, but with lower skill than local measurements. In contrast the North Atlantic Oscillation index does not correlate significantly with the surface EKE at intraseasonal and interannual scales. The analysis of altimeter data over the 1993–2001 supports the existence of this asymmetry between the regime locally forced by the atmosphere in the central basin, and the regime remotely forced by the incoming boundary current along the west Greenland coast. Those results have important implications for monitoring and predicting the surface eddy kinetic energy variability in the Labrador Sea.

Research highlights► We model the Labrador Sea circulation focusing on the surface eddy kinetic energy (EKE). ► We examine the processes responsible for the EKE interannual variability. ► In the West Greenland current region the surface current modulates the EKE. ► In the Central Labrador Sea (CLS) atmospheric fluxes modulate the EKE. ► A linear auto-regressive can predict the EKE in the CLS three months ahead.