Learning to Refine: Spectral-Decoupled Iterative Refinement Framework for Precipitation Nowcasting
This work addresses the critical need for accurate and physically consistent precipitation nowcasting for disaster mitigation, offering a deterministic solution that overcomes the blurring-hallucination trade-off.
Precipitation nowcasting suffers from a trade-off between regression models that produce over-smoothed predictions and diffusion models that generate unanchored hallucinations. The proposed SDIR framework reformulates nowcasting as progressive frequency-decoupled refinement, achieving significant improvements in spatial accuracy over SOTA methods while maintaining spectral fidelity competitive with diffusion-based approaches.
Accurate precipitation nowcasting is vital for disaster mitigation, but deep learning methods face a key trade-off: regression models produce over-smoothed, spectrally decaying predictions that blur convective details and violate turbulence power laws; diffusion models generate realistic yet unanchored hallucinations lacking physical grounding. We propose Spectral-Decoupled Iterative Refinement (SDIR), a deterministic framework that reformulates nowcasting as progressive frequency-decoupled refinement. SDIR first extracts a stable low-frequency synoptic skeleton, then iteratively refines high-frequency textures under physical constraints, eliminating both blurring and hallucinations. It features a dual-path design: the Synoptic Frequency-Guided Former (SFG-Former) with Scale-Adaptive Transformers for global structure, and the Fourier Residual Refiner (FR-Refiner) with Scale-Conditioned Fourier Neural Operators for fine residuals. A Physically Consistent Power Spectral Density (PCPSD) loss with dynamic masking enforces a turbulence-consistent spectral distribution. Experiments on three benchmarks show SDIR significantly outperforms SOTA methods in spatial accuracy while achieving spectral fidelity competitive with diffusion-based methods, enabling reliable high-resolution operational nowcasting. Code link: https://github.com/RuntimeWarning/SDIR.