Search papers, labs, and topics across Lattice.
The GeoDES model leverages a custom image-to-video diffusion approach to synthesize high-fidelity weather events, specifically targeting the evolving structures of cyclonic storms. This innovation addresses the limitations of existing regional and global weather models, which either lack detailed historical data or operate at insufficient resolutions. Evaluations reveal that GeoDES significantly enhances predictive accuracy, achieving a 52% reduction in Peak Vorticity Error and an 8% increase in Anomaly Correlation Coefficient compared to leading methods on the North Atlantic test set.
GeoDES achieves unprecedented accuracy in storm structure synthesis, outperforming existing models and redefining weather prediction capabilities.
While machine learning-based weather models hold significant promise, they struggle to predict the detailed structure of large-scale weather systems such as cyclonic storms. Regional models are constrained by limited historical records within fixed geographic boundaries, while global models are computationally expensive and often operate at resolutions too coarse to capture fine-grained storm dynamics. To bridge this gap, we introduce the Geospatial Diffusion-based Evolution Synthesis (GeoDES) model, a custom image-to-video diffusion model. By focusing generation strictly on the evolving storm structure, GeoDES synthesizes physically consistent, high-fidelity weather events suitable for stress-testing forecast models and expanding meteorological datasets. Evaluations demonstrate that GeoDES outperforms prior methods on key metrics, achieving $52\%$ lower Peak Vorticity Error and $8\%$ higher Anomaly Correlation Coefficient than the next strongest methods on the North Atlantic test set.
