Abstract
Interdecadal oscillations are analysed in a coupled ocean—atmosphere model made of a planetary geostrophic ocean model within an idealized geometry, coupled to a zonally-averaged tropospheric atmosphere model. The interdecadal variability that arises spontaneously in this coupled system is caused by intrinsic ocean dynamics, the coupled air-sea feedbacks being not essential. The spatial pattern of the variability bears some resemblance with observations and results obtained with atmosphere-ocean general circulation models (AOGCMs) as well as simpler climate models: large and quasi-stationary upper ocean temperature-dominated density anomalies are found in the north-western part of the ocean basin along with weaker, westward propagating anomalies in the remaining interior. The basic physical mechanism that lies at the heart of the existence of the interdecadal mode is a large-scale baroclinic instability of the oceanic mean flow in the vicinity of the western boundary, characteristic of ocean models forced by constant surface fluxes. Freshwater feedbacks associated with the hydrological cycle are found to have only a modest influence on the interdecadal mode. The presence of a periodic channel mimicking the Antarctic Circumpolar Current at high southern latitudes prevents the oceanic baroclinic instability to occur in the Southern Hemisphere.
Highlights
A proper detection of anthropogenic climate change requires a clear understanding of internal climate variability on timescales at which human influence on climate is most likely to occur (Houghton et al, 2001)
In the Atlantic Ocean, the spatial pattern associated to these long timescales is characterized by surface temperature anomalies of opposite sign between the Northern and Southern Hemispheres (NHs and SHs), suggesting a potential influence of variations in ocean heat transport associated with the large-scale Atlantic meridional overturning circulation (AMOC)
We have analysed interdecadal climate variability in a simplified coupled ocean–atmosphere model made of a 3-D flat-bottom planetary geostrophic ocean model and a 2-D zonally averaged climate model (ZACM) atmospheric model
Summary
A proper detection of anthropogenic climate change requires a clear understanding of internal climate variability on timescales at which human influence on climate is most likely to occur (Houghton et al, 2001). ∇ρ dominates the growth of density variance under FBCs where the ‘eddy part’ (prime) is defined as the departure from a timeaveraged (overbar) This term was associated with the long-wave limit of baroclinic instability in the vicinity of the western boundary current region by Colin de Verdiere and Huck (1999), later generalized by te Raa and Dijkstra (2002), to explain the interdecadal variability under FBCs. By contrast, the forcing term ρ B , where B represents the surface heat flux anomaly, dominates the growth of density variance under MBCs. The basic interpretation of this term led Arzel et al (2006) to conclude that a necessary condition to trigger variability under MBCs is that the correlation between temperature and salinity anomalies in the mixed layer must be positive and larger than the perturbation density ratio, which is a measure of the thermal against haline dependence of density.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
