Year: 2025
Type: article
Abstract: To this day, accurately simulating local-scale precipitation and reliably reproducing its distribution remains a challenging task. The limited horizontal resolution of global climate models is among the primary factors undermining their skill in this context. The physical mechanisms driving the onset and development of precipitation, especially in extreme events, operate at spatiotemporal scales smaller than those numerically resolved, thus struggling to be captured accurately. To circumvent this limitation, several downscaling approaches have been developed over the last decades to address the discrepancy between the spatial resolution of models’ output and the resolution required by local-scale applications. In this paper, we introduce RainScaleGAN, a conditional deep convolutional Generative Adversarial Network (GAN) for precipitation downscaling. GANs have been effectively used in image superresolution, an approach highly relevant for downscaling tasks. RainScaleGAN’s capabilities are tested in a perfect-model setup, where the spatial resolution of a precipitation dataset is artificially degraded from 0.25° × 0.25° to 2° × 2°, and RainScaleGAN is used to restore it. The developed model outperforms one of the leading precipitation downscaling methods found in the literature. RainScaleGAN not only generates a synthetic dataset featuring plausible high-resolution spatial patterns and intensities but also produces a precipitation distribution with statistics closely mirroring those of the ground-truth dataset. Given that RainScaleGAN’s approach is agnostic with respect to the underlying physics, the method has the potential to be applied to other physical variables such as surface winds or temperature. Significance Statement Accurately predicting local precipitation is difficult due to the limitations of current climate models. These models struggle to capture the small-scale processes causing precipitation, especially those leading to extreme events. To address this, we developed a new tool that uses advanced artificial intelligence techniques to improve rainfall predictions. This tool takes low-resolution precipitation data and enhances it to high resolution, providing more detailed rainfall patterns. Our results show that the tool performs better than one of the leading existing methods. This advancement could lead to more precise climate projections, better preparation for extreme weather, and suggests further exploration on additional weather variables.
Authors: Marcello Iotti, Paolo Davini, Jost von Hardenberg, Giuseppe Zappa
Language: English
Subfield: Atmospheric Science
Field: Earth and Planetary Sciences
Domain: Physical Sciences
SDG: Climate action
Open Access status: bronze
OpenAlex ID: W4410889451