Generative Adversarial Models for Extreme Downscaling of Climate Datasets

Addressing the challenges of climate change requires accurate and high-resolution mapping of climate and weather variables. However, many existing climate datasets, such as the gridded outputs of the state-of-the-art numerical climate models (e.g., general circulation models), are only available at very coarse spatial resolutions due to the model complexity and extremely high computational demand. Deep-learning-based methods, particularly generative adversarial networks (GANs) and their variants, have proved effective for refining natural images, and have shown great promise in improving scientific datasets. In this paper, we describe a conditional GAN-based geospatial downscaling method for extreme downscaling of gridded climate datasets. Compared to most existing methods, the method can generate high-resolution accurate climate datasets from very low-resolution inputs. More importantly, the method explicitly considers the uncertainty inherent to the downscaling process that tends to be ignored in existing methods. Given an input, the method can produce a multitude of plausible high-resolution samples instead of one single deterministic result. These samples allow for an empirical exploration and inferences of model uncertainty and robustness. With a case study of gridded climate datasets (wind velocity and solar irradiance), we demonstrate the performances of the framework in downscaling tasks with very high scaling factors (up to $64\times$) and highlight the advantages of the framework with a comprehensive comparison with commonly used downscaling methods, including area-to-point (ATP) kriging, deep image prior (DIP), enhanced deep super-resolution network (EDSR), enhanced super-resolution generative adversarial networks (ESRGAN), and physics-informed resolution-enhancing GAN (PhIRE GAN).