[2505.18179] GAIA: A Foundation Model for Operational Atmospheric Dynamics

Computer Science > Machine Learning arXiv:2505.18179 (cs) [Submitted on 15 May 2025 (v1), last revised 24 Mar 2026 (this version, v3)] Title:GAIA: A Foundation Model for Operational Atmospheric Dynamics Authors:Ata Akbari Asanjan, Olivia Alexander, Tom Berg, Stephen Peng, Jad Makki, Clara Zhang, Matt Yang, Disha Shidham, Srija Chakraborty, William Bender, Cara Crawford, Arun Ravindran, Olivier Raiman, David Potere, David Bell View a PDF of the paper titled GAIA: A Foundation Model for Operational Atmospheric Dynamics, by Ata Akbari Asanjan and 14 other authors View PDF HTML (experimental) Abstract:We introduce GAIA (Geospatial Artificial Intelligence for Atmospheres), a hybrid self-supervised geospatial foundation model that fuses Masked Autoencoders (MAE) with self-distillation with no labels (DINO) to generate semantically rich representations from global geostationary satellite imagery. Pre-trained on 15 years of globally-merged infrared observations (2001-2015), GAIA learns disentangled representations that capture atmospheric dynamics rather than trivial diurnal patterns, as evidenced by distributed principal component structure and temporal coherence analysis. We demonstrate robust reconstruction capabilities across varying data availability (30-95% masking), achieving superior gap-filling performance on real missing data patterns. When transferred to downstream tasks, GAIA consistently outperforms an MAE-only baseline: improving atmospheric river segmentation (F1: 0...