Contrastive Learning for Climate Model Bias Correction and Super-Resolution

TL;DR

This paper introduces a novel contrastive learning GAN approach for climate model bias correction and super-resolution, outperforming traditional methods by doubling spatial resolution and maintaining bias correction accuracy.

Contribution

It presents a new contrastive learning-based GAN method for climate data enhancement, addressing limitations of existing statistical bias correction techniques.

Findings

Achieves double the spatial resolution of NASA's climate product.

Maintains or improves bias correction accuracy for temperature and precipitation.

Enables more precise local climate hazard modeling.

Abstract

Climate models often require post-processing in order to make accurate estimates of local climate risk. The most common post-processing applied is bias-correction and spatial resolution enhancement. However, the statistical methods typically used for this not only are incapable of capturing multivariate spatial correlation information but are also reliant on rich observational data often not available outside of developed countries, limiting their potential. Here we propose an alternative approach to this challenge based on a combination of image super resolution (SR) and contrastive learning generative adversarial networks (GANs). We benchmark performance against NASA's flagship post-processed CMIP6 climate model product, NEX-GDDP. We find that our model successfully reaches a spatial resolution double that of NASA's product while also achieving comparable or improved levels of bias correction in both daily precipitation and temperature. The resulting higher fidelity simulations of present and forward-looking climate can enable more local, accurate models of hazards like flooding, drought, and heatwaves.