Statistical Inference for Stochastic Process Models in Weather and Climate Science

Stochastic process models occur in many areas of weather and climate science. They are used in atmosphere and ocean science at different time scales ranging from large-scale atmospheric flow at weather time scales, to interannual phenomena such as the El Niño Southern Oscillation (ENSO), to interdecadal variability. In palaeoclimate such process models have been used for rapid glacial climate transitions, the so-called Dansgaard-Oeschger events, or for glacial-interglacial cycles. Motivation for deriving inverse models may be manifold: They are used for predictive purposes and risk assessment, they may serve to diagnose or interpret observational data or data from complex physics-based models; or they may be meant to extract particular features or processes with the goal to understand the dynamical mechanism behind these. The workshop aims at bringing together weather and climate scientists, on one hand, and statisticians, on the other, specialising in the use of mathematical and statistical techniques for inference in stochastic processes, random dynamical systems and time series modelling. Principal goals of the workshop are exchange of ideas and information between the two groups on kinds of problems and challenges they face in their research, and techniques they employ for their solution. More specifically, the emphasis of the workshop is on a model- and data-driven, statistical approach to climate and weather science, and the use of modern statistical techniques for extraction of model structures and features from the noisy and incomplete data, to be used subsequently for explanatory or predictive purposes.