Simulation of extreme events and rare transitions in climate models with rare event algorithms

The analysis of extreme events in today's and future climate is an important area of application of numerical climate models. Extremes like heat waves, floods and wind storms, or rare abrupt transitions associated with tipping elements of the climate system, can have severe impacts on human societies and ecosystems. Studying these events on a robust statistical basis with complex numerical models is however computationally challenging, as very long simulations and/or very large ensembles are necessary to sample a sufficient number of events to have acceptable levels of statistical accuracy. This problem can be tackled using rare event algorithms, numerical tools designed to reduce the computational effort required to sample rare events in numerical models. These methods typically take the form of genetic algorithms, where a set of suppression and cloning rules are applied to the members of an ensemble simulation, in order to oversample trajectories leading to the events of interest. In this talk I will show recent applications of these methods to different classes of events, fousing in particular on heat waves and warm summers in the Northern hemisphere, and on extremes of Arctic sea ice reduction. I will then show how rare event algorithms can be used to simulate abrupt transitions associated to the weakening and collapse of the Atlantic Meridional Overturning Circulation induced purely by internal variability. Finally I will discuss the relevance of these techiniques for future applications, in particular for seasonal to decadal predictions, the analysis of tipping points and the validation of early warning indicators.