DiffSR: Learning Radar Reflectivity Synthesis via Diffusion Model from Satellite Observations
This addresses the need for accurate radar data in areas with missing ground observations, particularly for convective weather monitoring, though it is an incremental improvement over existing methods.
The paper tackles the problem of over-smoothing in weather radar data synthesis from satellite observations by proposing DiffSR, a two-stage diffusion-based method that generates high-frequency details and high-value areas, achieving state-of-the-art results.
Weather radar data synthesis can fill in data for areas where ground observations are missing. Existing methods often employ reconstruction-based approaches with MSE loss to reconstruct radar data from satellite observation. However, such methods lead to over-smoothing, which hinders the generation of high-frequency details or high-value observation areas associated with convective weather. To address this issue, we propose a two-stage diffusion-based method called DiffSR. We first pre-train a reconstruction model on global-scale data to obtain radar estimation and then synthesize radar reflectivity by combining radar estimation results with satellite data as conditions for the diffusion model. Extensive experiments show that our method achieves state-of-the-art (SOTA) results, demonstrating the ability to generate high-frequency details and high-value areas.