Bridging physical, analytical, information-theoretic and machine learning approaches to system dynamics and predictability across Hydrology and Earth System Sciences
Convener:
Rui A. P. Perdigão
|
Co-conveners:
Julia Hall,
Mohammad Azizur Rahman,
Maria KireevaECSECS,
Cristina Prieto
Special focus is given to unveil complex system dynamics, regimes, transitions, extremes, hazards and their interactions, along with their physical understanding, predictability and uncertainty, across multiple spatiotemporal scales.
The session encourages discussion on interdisciplinary physical and data-based approaches to system dynamics across Hydrology and broader Geosciences, ranging from novel advances in stochastic, computational, information-theoretic and dynamical system analysis, to cross-cutting emerging pathways in information physics and systems intelligence.
The session further encompasses practical aspects of working with systems intelligence and information theoretic approaches for model evaluation and uncertainty analysis, causal inference and process networks, along with hydrological and geophysical automated learning, model design, prediction and decision support.
Contributions are gathered from an interdisciplinary community working with diverse approaches ranging from dynamical modelling to data mining, machine learning, artificial intelligence and beyond along with their interconnections with physical process understanding in mind.
Take part in a thrilling session exploring and discussing promising avenues in system dynamics and information discovery, quantification, modelling and interpretation, where methodological ingenuity and natural process understanding come together to shed light onto fundamental theoretical aspects to build innovative methodologies to tackle real-world challenges facing our planet.