Extreme weather, from hurricanes to atmospheric rivers, increasingly disrupts lives and infrastructure. To improve preparedness and mitigation, this session explores the integration of AI/ML with traditional numerical weather prediction (NWP). While NWP provides a vital mechanistic understanding of the atmosphere, AI offers faster computation and advanced pattern recognition in complex datasets. Rather than viewing these as competing paradigms, we focus on hybrid modeling and physics-informed machine learning. Key topics include ML-based downscaling, data assimilation, model emulation, and operational hybrid systems. We will also address critical challenges like data scarcity, model interpretability, and uncertainty quantification. We invite researchers and practitioners from atmospheric science, data science, and climate adaptation to join this cross-disciplinary dialogue. Our goal is to advance a scientifically robust and societally relevant framework for next-generation extreme weather prediction.

• H - Hydrology
• NH - Natural Hazards