A 20-year coupled ocean-sea ice-atmosphere variability mode in the North Atlantic in an AOGCM

• Main Authors: Escudier, R., Mignot, Juliette, Swingedouw, D.
• Format: Text
• Language: English
• Published: 2013
• Subjects: Decadal climate variability; Thermohaline circulation; Ocean-sea ice-atmosphere coupling; Subpolar North Atlantic; Coupled climate model; Greenland; Denmark Strait; East Greenland; east greenland current; Labrador Sea; Nordic Seas; North Atlantic; Sea ice
• Online Access: http://www.documentation.ird.fr/hor/fdi:010058989

In order to understand potential predictability of the ocean and climate at the decadal time scales, it is crucial to improve our understanding of internal variability at this time scale. Here, we describe a 20-year mode of variability found in the North Atlantic in a 1,000-year pre-industrial simulation of the IPSL-CM5A-LR climate model. This mode involves the propagation of near-surface temperature and salinity anomalies along the southern branch of the subpolar gyre, leading to anomalous sea-ice melting in the Nordic Seas, which then forces sea-level pressure anomalies through anomalous surface atmospheric temperatures. The wind stress associated to this atmospheric structure influences the strength of the East Greenland Current across the Denmark Strait, which, in turn, induces near-surface temperature and salinity anomalies of opposite sign at the entrance of the Labrador Sea. This starts the second half of the cycle after approximatively 10 years. The time scale of the cycle is thus essentially set by advection of tracers along the southern branch of the subpolar gyre, and by the time needed for anomalous East Greenland Current to accumulate heat and freshwater anomalies at the entrance of the Labrador Sea. The Atlantic meridional overturning circulation (AMOC) does not play a dominant role in the mode that is confined in the subpolar North Atlantic, but it also has a 20-year preferred timescale. This is due to the influence of the propagating salinity anomalies on the oceanic deep convection. The existence of this preferred timescale has important implications in terms of potential predictability of the North Atlantic climate in the model, although its realism remains questionable and is discussed.