Please note that this session was withdrawn and is no longer available in the respective programme. This withdrawal might have been the result of a merge with another session.
NP2.1 | Stochastic Dynamics, Uncertainty Quantification and Probabilistic Machine Learning in the Earth Sciences
EDI
Stochastic Dynamics, Uncertainty Quantification and Probabilistic Machine Learning in the Earth Sciences
Convener: Balasubramanya Nadiga | Co-conveners: Paul Williams, Christian Franzke, Naiming Yuan, Hannah ChristensenECSECS
Since key earth system processes are often stochastic (besides also being nonlinear and multiscale), it is important to avoid a false sense of confidence that arises from a modeling perspective that ignores the probabilistic nature of such processes. This is indeed also the case when machine learning is used to model such processes, as has become commonplace in the last decade. In the context of modeling such processes, this session seeks to explore, understand and promote interactions between specialized approaches such as first-principles based modeling, dynamical systems theory, complex systems theory, complex network theory, etc. on the one hand and general approaches such as machine learning and probabilistic machine learning approaches on the other hand.

We seek contributions on all aspects of stochastic dynamics in the various components of the Earth system such as the atmosphere, the ocean, the cryosphere, the solid earth, etc. and in the various coupled processes such as the hydrological cycle, the biogeochemical cycles, etc. and climate. Communications on theoretical, experimental and modeling studies are all welcome, where the latter modeling studies can span the range of model hierarchy from idealized models to complex Earth System Models (ESM) and use either traditional or machine learning based approaches.