Saved in:
Bibliographic Details
Main Authors: Delefosse, Aymeric, Charantonis, Anastase, Béréziat, Dominique
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.00897
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914437392236544
author Delefosse, Aymeric
Charantonis, Anastase
Béréziat, Dominique
author_facet Delefosse, Aymeric
Charantonis, Anastase
Béréziat, Dominique
contents Machine learning-based weather forecasting models now surpass state-of-the-art numerical weather prediction systems, but training and operating these models at high spatial resolution remains computationally expensive. We present a modular framework that decouples forecasting from spatial resolution by applying learned generative super-resolution as a post-processing step to coarse-resolution forecast trajectories. We formulate super-resolution as a stochastic inverse problem, using a residual formulation to preserve large-scale structure while reconstructing unresolved variability. The model is trained with flow matching exclusively on reanalysis data and is applied to global medium-range forecasts. We evaluate (i) design consistency by re-coarsening super-resolved forecasts and comparing them to the original coarse trajectories, and (ii) high-resolution forecast quality using standard ensemble verification metrics and spectral diagnostics. Results show that super-resolution preserves large-scale structure and variance after re-coarsening, introduces physically consistent small-scale variability, and achieves competitive probabilistic forecast skill at 0.25° resolution relative to an operational ensemble baseline, while requiring only a modest additional training cost compared with end-to-end high-resolution forecasting.
format Preprint
id arxiv_https___arxiv_org_abs_2604_00897
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Super-Resolving Coarse-Resolution Weather Forecasts With Flow Matching
Delefosse, Aymeric
Charantonis, Anastase
Béréziat, Dominique
Machine Learning
Computer Vision and Pattern Recognition
Machine learning-based weather forecasting models now surpass state-of-the-art numerical weather prediction systems, but training and operating these models at high spatial resolution remains computationally expensive. We present a modular framework that decouples forecasting from spatial resolution by applying learned generative super-resolution as a post-processing step to coarse-resolution forecast trajectories. We formulate super-resolution as a stochastic inverse problem, using a residual formulation to preserve large-scale structure while reconstructing unresolved variability. The model is trained with flow matching exclusively on reanalysis data and is applied to global medium-range forecasts. We evaluate (i) design consistency by re-coarsening super-resolved forecasts and comparing them to the original coarse trajectories, and (ii) high-resolution forecast quality using standard ensemble verification metrics and spectral diagnostics. Results show that super-resolution preserves large-scale structure and variance after re-coarsening, introduces physically consistent small-scale variability, and achieves competitive probabilistic forecast skill at 0.25° resolution relative to an operational ensemble baseline, while requiring only a modest additional training cost compared with end-to-end high-resolution forecasting.
title Super-Resolving Coarse-Resolution Weather Forecasts With Flow Matching
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2604.00897