RainScaler: A Physics-Inspired Network for Precipitation Correction and Downscaling

Spatial downscaling of precipitation, in which fine-grained regional precipitation patterns are recovered from coarse-resolution images, plays a crucial role in various weather and meteorological analyses. However, the intricate noise information presented in the observation data intertwines with the fine-scale characteristics, which poses challenges for subsequent feature extraction. Regional precipitation suffers from complex spatial patterns. Moreover, the real observatory data contains information inconsistent with the established physical principle, due either to inaccurate or incomplete physical models or limited data quality, thus making the implementation of physically informed deep learning (DL) more difficult. For example, strong physical constraints may lead to over-regularization, in which the model becomes too rigid and fails to capture certain complexities in the data. In this work, we propose RainScaler, a physics-inspired deep neural network, to tackle these issues. First, to remove the noise and preserve the vital precipitation patterns effectively, the proposed RainScaler exploits an inconsistency-aware (IA) Denoising Net to explicitly model the spatial variability of noise in the input. In addition, a graph module is designed to learn the geographical-dependent fine-grained patterns in high-dimensional feature space at a moderate computation cost. Finally, multiscale physical constraints are skillfully embedded to incorporate additional insights into the data-driven framework. We test our approach on a public dataset consisting of over 60000 real low-resolution (LR) and high-resolution (HR) precipitation map pairs collected by different sensors. Our method produces realistic-looking precipitation maps with better discernment capability and corrects the structural error of precipitation distribution, especially for extreme events. Moreover, we evaluate the potential risks of incorporating physical constraints in real-world data applications. Our method unveils opportunities for multisource data fusion and provides possible solutions to improve the physical feasibility of data-driven models.