Recent advances in Geospatial AI (GeoAI) are transforming how multi-source Earth observation (EO) data are integrated and used for environmental monitoring. This session invites contributions that use machine learning and data-driven approaches to integrate satellite, in situ, and model-based datasets to improve understanding of Earth system processes. Topics of interest include, but are not limited to, multi-source data fusion (e.g., SAR, optical, LiDAR), data integration with machine learning, digital twins of Earth systems, and cross-scale environmental monitoring. We particularly encourage submissions that consider uncertainty, ensure reliability in noisy or incomplete observations, and develop scalable, operational GeoAI frameworks. Applications may include climate extremes, ocean and coastal monitoring, land-use/cover change, and disaster risk monitoring and management. This session aims to bridge methodological advances and real-world environmental challenges, and to promote interdisciplinary collaboration across the Earth and data sciences.

• IN - Informatics
• NH - Natural Hazards