The increasing availability of process-based models, data-driven methods, and hybrid approaches incorporating AI tools is transforming hydrology. These simulation and prediction frameworks can integrate diverse data sources, including in-situ measurements, high resolution satellite observations, and geophysical surveys, to improve uncertainty quantification and predictive skill of hydrologic models.

This session invites contributions on theory advancement and/or applications of advanced AI techniques (e.g., physics-informed machine learning, agentic AI, generative AI, hybrid differentiable modeling, etc.) and statistical methods (i.e., inverse modeling, Bayesian inference, data assimilation, etc.) for hydrologic and hydrometeorological uncertainty quantification. We particularly welcome studies that integrate multiple datasets for model calibration or validation, perform large-domain parameter estimation, or couple AI with data assimilation algorithms to address surface water-groundwater interactions and hydrologic extremes.