{"ID":2866555,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.21399","arxiv_id":"2509.21399","title":"Downscaling climate projections to 1 km with single-image super resolution","abstract":"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.","short_abstract":"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-...","url_abs":"https://arxiv.org/abs/2509.21399","url_pdf":"https://arxiv.org/pdf/2509.21399v2","authors":"[\"Petr Košťál\",\"Pavel Kordík\",\"Ondřej Podsztavek\"]","published":"2025-09-24T12:19:51Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.LG\"]","methods":"[]","has_code":false}