Aerial photograph of sea ice in the Arctic Ocean, one of the triggers of the North Atlantic thermohaline circulation, taken from a commercial flight over the coast of Svarlbard/Spitzbergen in mid-April 2009 (Photo: © 2009 Martin Beniston).


CADMOS, the Center for Advanced Modeling Sciences, brings together the University of Geneva, the University of Lausanne, and the Swiss Federal Institute of Technology in Lausanne (EPFL) to focus on complex issues in the natural and engineering sciences, as well as economics and social issues, using High Performance Computing (HPC). The principal machine at the disposal of the three universities is the IBM "Blue Gene" supercomputer.

The contribution of our Institute and Research Group to CADMOS is through the study of non-linearities and bifurcations in the climate system. Led by Dr. Maura Brunetti, a particular focus is given to ocean-atmosphere, sea ice-atmosphere-ocean couplings that are believed to be responsible for much of the observed climate variability and decadal-to-century scale oscillations such as the NAO (North Atlantic Oscillation).

In this context we investigate the critical role of the North Atlantic meridional overturning circulation on the high-latitude climate and in particular on the temperate climate of Northwest Europe. Paleoclimate reconstructions have shown that this overturning cell can change its operation mode (from `ON' to `OFF' state) on decadal time scales, giving rise to significant climate impacts. The stability properties of the Atlantic Meridional overturning circulation have been investigated through climate numerical models of different complexity. They generally confirm the existence of switches between ON and OFF modes, which depend on outbursts of freshwater into the Atlantic ocean. However, the physical processes which effectively determine the state of the Atlantic Meridional overturning circulation are not yet fully understood.

The CADMOS project aims to investigate the freshwater budget of the Atlantic basin using an intermediate complexity code, the MIT general circulation model in the configuration where ocean and sea-ice dynamics are coupled to simplified descriptions of atmosphere and land. We compare different stability indicators proposed in the literature and discuss how the salinity content affects the behaviour of the North Atlantic ocean circulation. In particular, we investigate nonlinear dependences of the ON/OFF mode on the averaged salinity content in the Atlantic ocean.


The MIT ocean-atmosphere model code is the principal modeling tool currently used in the context of CADMOS at UNIGE