Snow science presents a fundamental challenge for artificial intelligence because snowpack evolution emerges from strongly coupled physical processes governing accumulation, metamorphism, melt, runoff, and land-atmosphere interactions across multiple scales. This session invites contributions on physics-informed AI approaches that integrate machine learning with physical constraints, process understanding, and Earth system knowledge to improve snow prediction and scientific discovery. Topics include not limited to snow water equivalent (SWE), snow data, hybrid physics-AI, remote sensing and foundation models, uncertainty quantification, and AI methods for noisy, incomplete, or scale-mismatched snow observations.

We also encourage contributions addressing process representation, multiscale coupling, mountain hydrology, snow-climate interactions, and the integration of AI with observational, reanalysis, and model-based snow information. Contributions exploring how physics-informed AI can improve prediction across regions, support water supply forecasting, and reveal new understanding of snow processes under changing climate conditions are particularly welcome.

• GC - Global Environmental Change
• H - Hydrology
• IN - Informatics