Scientific objectives

Scientific objectives

The first objective of the project (WP1) is to embed the analogue method in the theory of dynamical systems in order to be able to provide a metric of an attractor deformation in time. The groundbreaking issue of WP1 is to build a methodology to qualify the attractor changes (translation or bifurcation) of the climate system, without an assumption of stationarity. The main achievement will be to derive probabilistic estimates of the quality of analogues and estimate the times of change from observed trajectories. This methodological step is important and has been overlooked in the literature. It will shed a new light to theoretical climate dynamics, by providing the necessary statistical assessments.

We will create an open source computer toolkit to compute flow analogues from a wide array of databases (WP2). This toolkit will be the cornerstone of the two climate applications we envisage. It will be distributed to the scientific community for a wide use and other applications. It is aimed at becoming a standard for this type of analysis, since we have identified a large demand for such toolkit “products”, especially for the creation of climate services.

We have identified two major challenges in climate/atmospheric science related to climate change (WP3): How the atmospheric variability is altered in the northern extra-tropics when an external forcing is applied (solar, volcanic and anthropogenic), How the probability of extreme climate events depends on climate forcings.

These challenges will be tackled with the methodology of flow analogues. The second challenge will implement a pseudo real-time analysis of extreme events. By analyzing sequences of extreme events (heat and cold spell, droughts and heavy precipitation), we will be able to estimate how slow climate forcings affect meteorological variability. This has wide implications for the evaluation of climate change risks in several regions of the world.

Analogue methods have mainly been applied on time-limited datasets, which sample a fraction of the phase space. The breakthrough of the project is to generalize this methodology for a use with large ensembles of observations and simulations. In this way, this project will bridge a gap between operational needs (i.e. the immediate analysis of climate events) and the understanding long-term climate changes, by elaborating and validating a versatile statistical methodology of flow analogues.