SwAIther-Precip: Lead-Time-Aware Bias Correction Enables Kilometer-Scale Downscaling of Global AI Precipitation Forecasts over Switzerland
Researchers have developed SwAIther-Precip, a new framework designed to improve the resolution and accuracy of AI-driven precipitation forecasts. This method specifically addresses biases in global AI weather models that are dependent on the forecast lead time. By correcting these biases before applying a diffusion-based super-resolution model, SwAIther-Precip can generate kilometer-scale precipitation fields with significantly improved accuracy and spatial fidelity. AI
IMPACT Enhances the utility of global AI weather models for localized, high-resolution precipitation forecasting.