A model study of decadal climate variability and predictability associated with the Atlantic meridional overturning circulation

Abstract

This study addresses the decadal variability and predictability of the Atlantic Meridional Overturning Circulation (AMOC), and associated key variables, in two IPCC-class climate models. The AMOC variability is analyzed in a new climate model CHIME, which features a novel (largely isopycnic) ocean component. Power Spectral analysis reveals enhanced variability for periods in the range 15-30 years. The primary mode of variability is associated with decadal changes in the Labrador and the Greenland-Iceland-Norwegian (GIN) seas, in both cases linked to the tropical activity about 15 years earlier. These decadal changes are controlled by the low-frequency North Atlantic Oscillation (NAO), associated with a tropical-extratropical teleconnection. Poleward advection of salinity anomalies in the mixed layer also leads to AMOC changes that are linked to convective processes in the Labrador Sea. A secondary mode of variability is associated with interannual changes in the Labrador and GIN Seas, through the impact of the NAO on local surface density. The decadal potential predictability of the AMOC and climate as represented in the non-isopycnic IPSL-CM5A model and CHIME is explored using prognostic and diagnostic approaches. The modelled AMOC has an average predictive skill of 8 and 6 years, respectively. Over the ocean, surface temperature has the highest skill up to 2 decades in the far north of the North Atlantic, in both models. Additional oceanic areas of predictability are identified in IPSL-CM5A in the tropics and subtropics. The spatio-temporal predictability of both surface temperature over land and precipitation differs somewhat between the two models, but is of limited extent compared to that of ocean variables. Predictability of climate arises from the mechanisms controlling the decadal AMOC fluctuations. Predictive skills of AMOC and climate are favoured by extreme AMOC events but the role of minimum versus maximum states remains to be clarified. The expected better predictive skills of CHIME over non-isopycnic models (due to its better preservation of water masses and more coherent internal structure to the anomalies) are not borne out.

More information

Identifiers

Catalogue record

Export record