A Likelihood-Based Generative Approach for Spatially Consistent Precipitation Downscaling

TL;DR

This paper introduces a novel likelihood-based generative method for precipitation downscaling that combines likelihood and adversarial losses to improve spatial consistency in projections.

Contribution

It proposes a new approach that fuses likelihood-based and adversarial losses, addressing spatial inconsistency issues in precipitation downscaling models.

Findings

Enhanced spatial consistency in precipitation projections

Combines strengths of likelihood and adversarial training

Improves accuracy of downscaled precipitation data

Abstract

Deep learning has emerged as a promising tool for precipitation downscaling. However, current models rely on likelihood-based loss functions to properly model the precipitation distribution, leading to spatially inconsistent projections when sampling. This work explores a novel approach by fusing the strengths of likelihood-based and adversarial losses used in generative models. As a result, we propose a likelihood-based generative approach for precipitation downscaling, leveraging the benefits of both methods.