This session focuses on the emerging role of artificial intelligence (AI) in ocean inverse problems, data assimilation, and scientific discovery. We invite contributions that integrate AI with physics-based inference to improve estimation of ocean states, parameters, fluxes, sources, and processes across physical, biogeochemical, ecological, and coupled ocean systems. Topics of interest include Bayesian and variational inverse methods, hybrid AI-data assimilation frameworks, learned observation operators and priors, surrogate and emulators for efficient inference, uncertainty quantification, sensor fusion, and interpretable machine learning for understanding ocean processes. We particularly welcome studies spanning both fundamental ocean science and relevant applications, including ocean forecasting, climate and ecosystem dynamics, biogeochemical cycles, marine pollution, and coastal processes. By bridging inverse theory, data assimilation, and AI, this session aims to highlight new pathways for turning sparse and complex ocean observations into robust predictions, mechanistic understanding, and actionable knowledge.

• NG - Nonlinear Geophysics