The integration of machine learning (ML), physical principles, and hybrid modeling approaches is rapidly advancing hydrology and river system modeling. These developments are enhancing predictive skill, improving interpretability, and enabling new pathways for data-driven and process-informed scientific discovery across scales. This session invites contributions on: (1) Data-driven and hybrid ML models; (2) Physics-informed and learnable physical models; (3) Scalable ML approaches from watershed to global scales; (4) Transferable and large-scale modeling frameworks across regions and conditions; (5) AI-integrated Earth system modeling; (6) Discovery of hydrologic patterns and relationships using ML; and (7) Trustworthy, interpretable, and responsible AI. We particularly welcome contributions that bridge data and physical understanding, improve model generalization, and advance robust, scalable, and physically consistent representations of hydrologic systems.

• IN - Informatics