Accurate weather forecasts are crucial to planning disaster management, protecting existing infrastructure, and sustaining federal mission areas.
Geospatial foundation models (GFMs) have recently been developed as impactful tools for various stakeholders as vision transformers increasingly outperform numerical weather prediction (NWP) models. The state-of-the-art for machine learning-based weather foundation models has been recently fueled by the vision transformer architecture and may be finetuned for regional or local weather predictions.
We demonstrate promising results on smaller vision transformers and propose that this method improves performance on regional finetuning tasks in the vicinity of the 180° line of longitude.
Published on ESS Open Archive: The Earth and Space Science Open Archive is a community server established to accelerate the open discovery and dissemination of earth, environmental, and space science research by archiving and sharing early research outputs, including preprints, presentations from major scientific meetings, and important documents of scholarly societies.
DOI: 10.22541/essoar.173463107.72488027/v1