Downscaling climate projections to 1 km with single-image super resolution

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Košťál, Petr, Kordík, Pavel, Podsztavek, Ondřej
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918230690365440
author Košťál, Petr
Kordík, Pavel
Podsztavek, Ondřej
author_facet Košťál, Petr
Kordík, Pavel
Podsztavek, Ondřej
contents High-resolution climate projections are essential for local decision-making. However, available climate projections have low spatial resolution (e.g. 12.5 km), which limits their usability. We address this limitation by leveraging single-image super-resolution models to statistically downscale climate projections to 1-km resolution. Since high-resolution climate projections are unavailable, we train models on a high-resolution observational gridded data set and apply them to low-resolution climate projections. We cannot evaluate downscaled climate projections with common metrics (e.g. pixel-wise root-mean-square error) because we lack ground-truth high-resolution climate projections. Therefore, we evaluate climate indicators computed at weather station locations. Experiments on daily mean temperature demonstrate that single-image super-resolution models can downscale climate projections without increasing the error of climate indicators compared to low-resolution climate projections.
format Preprint
id arxiv_https___arxiv_org_abs_2509_21399
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Downscaling climate projections to 1 km with single-image super resolution
Košťál, Petr
Kordík, Pavel
Podsztavek, Ondřej
Computer Vision and Pattern Recognition
Machine Learning
High-resolution climate projections are essential for local decision-making. However, available climate projections have low spatial resolution (e.g. 12.5 km), which limits their usability. We address this limitation by leveraging single-image super-resolution models to statistically downscale climate projections to 1-km resolution. Since high-resolution climate projections are unavailable, we train models on a high-resolution observational gridded data set and apply them to low-resolution climate projections. We cannot evaluate downscaled climate projections with common metrics (e.g. pixel-wise root-mean-square error) because we lack ground-truth high-resolution climate projections. Therefore, we evaluate climate indicators computed at weather station locations. Experiments on daily mean temperature demonstrate that single-image super-resolution models can downscale climate projections without increasing the error of climate indicators compared to low-resolution climate projections.
title Downscaling climate projections to 1 km with single-image super resolution
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2509.21399