Extreme Precipitation Nowcasting using Multi-Task Latent Diffusion Models
This work addresses the problem of improved extreme precipitation nowcasting for meteorology and disaster management, representing a novel method for a known bottleneck.
The paper tackles the challenge of accurately capturing spatial details in precipitation radar images, especially for high-intensity regions, by proposing a Multi-Task Latent Diffusion Model (MTLDM) that decomposes images by precipitation intensity and integrates them for prediction, resulting in a 13-26% improvement in Critical Success Index (CSI) over state-of-the-art methods.
Deep learning models have achieved remarkable progress in precipitation prediction. However, they still face significant challenges in accurately capturing spatial details of radar images, particularly in regions of high precipitation intensity. This limitation results in reduced spatial localization accuracy when predicting radar echo images across varying precipitation intensities. To address this challenge, we propose an innovative precipitation prediction approach termed the Multi-Task Latent Diffusion Model (MTLDM). The core idea of MTLDM lies in the recognition that precipitation radar images represent a combination of multiple components, each corresponding to different precipitation intensities. Thus, we adopt a divide-and-conquer strategy, decomposing radar images into several sub-images based on their precipitation intensities and individually modeling these components. During the prediction stage, MTLDM integrates these sub-image representations by utilizing a trained latent-space rainfall diffusion model, followed by decoding through a multi-task decoder to produce the final precipitation prediction. Experimental evaluations conducted on the MRMS dataset demonstrate that the proposed MTLDM method surpasses state-of-the-art techniques, achieving a Critical Success Index (CSI) improvement of 13-26%.