AirCast-SR: A Foundation Model for Kilometer-Scale Atmospheric Super-Resolution via Latent Consistency Diffusion

AirCast-SR, a new foundation model for atmospheric super-resolution, downscales global AI weather forecasts from 28-kilometer to 1-kilometer horizontal resolution at hourly intervals, producing 67-hour forecasts of eight surface variables. The model, which uses a three-dimensional U-Net within a Latent Consistency Model diffusion framework, achieves near-zero bias and preserves fine-scale atmospheric structure across wavelengths from 10 to 100 kilometers. Validated over the contiguous United States and demonstrating zero-shot global transferability to India and Germany, AirCast-SR establishes a new paradigm for kilometer-scale AI weather prediction for applications in energy, agriculture, and disaster management.

arXiv:2605.26130v1 Announce Type: new Abstract: Operational weather prediction at kilometer scales remains computationally prohibitive for traditional numerical weather prediction NWP models, limiting forecast access for applications in energy, agriculture, and disaster management that require fine-grained spatiotemporal detail. Here we introduce AirCast-SR, a foundation model for atmospheric super-resolution that downscales global AI weather forecasts from 0.25 degree ~28 km to 1 km horizontal resolution at hourly temporal resolution, producing 67-hour forecasts of eight coupled surface variables simultaneously. EarthMind-SR employs a three-dimensional U-Net conditioned within a Latent Consistency Model LCM diffusion framework, trained on patch-based samples over the contiguous United States CONUS using GraphCast forecasts as input and NOAA's Analysis of Record for Calibration AORC as the target. The model achieves near-zero bias across all variables and lead times, and its radial power spectral density analysis demonstrates preservation of fine-scale atmospheric structure at wavelengths of 10 km to 100 km where coarser models lose spectral power. We validate EarthMind-SR across three CONUS case studies spanning winter, summer, and spring seasons, and demonstrate zero-shot global transferability over India and Germany using independent surface station observations without any retraining or fine-tuning. As an open-weights foundation model, EarthMind-SR establishes a new paradigm for kilometer-scale AI weather prediction and provides a platform for regional fine-tuning, distillation, and downstream applications in climate services and hazard forecasting.