A Differentiable Framework for Global Circulation Model Precipitation Bias Correction

TL;DR

This paper introduces dCLIMBA, a differentiable, adaptive bias correction framework for GCM precipitation outputs, improving accuracy in extreme events and spatial patterns, with better generalization than traditional methods.

Contribution

The study presents a novel differentiable bias-adjustment method that learns spatiotemporal correction patterns, enhancing precipitation bias correction across diverse regions.

Findings

Corrects extreme precipitation magnitude and distribution effectively.

Reproduces precipitation quantile distribution across U.S. cities.

Shows partial preservation of future trends and reduces biases in unseen regions.

Abstract

Systematic biases in General Circulation Model (GCM) outputs limit their direct applicability in regional planning, making bias correction a technically demanding but necessary step for both short-term and long-term impact assessment. Correcting precipitation is particularly challenging due to its non-Gaussian distribution, intermittent nature, and heavy-tailed extremes. However, traditional statistical bias-correction methods have limited ability to learn systematic patterns from large datasets or generalize to new locations. While machine learning (ML) provides greater flexibility, it can produce unpredictable and difficult-to-interpret results, limiting generalization across GCMs and locations. In this study, we propose a differentiable bias-adjustment framework called dCLIMBA, that learns a spatiotemporally adaptive parametric bias-adjustment procedure, rather than corrected precipitation directly, between historical CMIP6 model outputs and a gridded observation-based dataset, Livneh. Results demonstrate that the proposed method corrects the magnitude and distribution of extreme precipitation with particularly strong performance in the upper tail. The quantile distribution of precipitation was well reproduced across diverse U.S. cities, and spatial patterns were comparable to those from the widely used LOCA2 statistical downscaling product. In addition, the framework showed partial future trend preservation and promising attenuation of marginal biases in unseen regions. This work presents a modular and efficient bias-correction approach. The differentiable approach provides an easy-to-use option for connecting atmospheric-model outputs to on-the-ground impacts.