Detecting and classifying bifurcations in the climate system

We propose to develop a general method for studying bifurcations and transitions of the Earth system, with particular focus on global climate change. As the states of the climate system can be described by analytic potential, we propose a study in which the potential would be estimated from an observed record of a climate variable. Knowing the potential and recently observed trends in the climate system (in temperature, sea level, etc.), we plan to perform computer simulations and evaluate the changes the climate potentials undergo. We will apply the bifurcation theory to study the evolution of the potential and possible bifurcations of its extremals. This will allow to make projections of time series which will be compared with projections provided by IPCC AR4. Once the method finally developed, it will be tested on artificial data with simulated dynamics, and then applied on available paleo-, historical and real-time observed data.