Academic Journal
Benchmarking the geographic generalization of deep learning models for precipitation downscaling.
| Τίτλος: | Benchmarking the geographic generalization of deep learning models for precipitation downscaling. |
|---|---|
| Συγγραφείς: | Harder P; Mila - Quebec AI Institute, Montreal, Canada.; European Centre for Medium Range Weather Forecasts (ECMWF), Bonn, Germany., Schmidt L; Mila - Quebec AI Institute, Montreal, Canada. luca-marie.schmidt@mila.quebec.; Cluster of Excellence Machine Learning, University of Tübingen, Tübingen, Germany. luca-marie.schmidt@mila.quebec., Pelletier F; Mila - Quebec AI Institute, Montreal, Canada., Ludwig N; Cluster of Excellence Machine Learning, University of Tübingen, Tübingen, Germany., Chantry M; European Centre for Medium Range Weather Forecasts (ECMWF), Bonn, Germany., Lessig C; European Centre for Medium Range Weather Forecasts (ECMWF), Bonn, Germany., Hernandez-Garcia A; Mila - Quebec AI Institute, Montreal, Canada.; Université de Montréal, Montreal, Canada., Rolnick D; Mila - Quebec AI Institute, Montreal, Canada.; McGill University, Montreal, Canada. |
| Πηγή: | Scientific reports [Sci Rep] 2026 Jan 27; Vol. 16 (1), pp. 3733. Date of Electronic Publication: 2026 Jan 27. |
| Τύπος έκδοσης: | Journal Article |
| Γλώσσα: | English |
| Στοιχεία περιοδικού: | Publisher: Nature Publishing Group Country of Publication: England NLM ID: 101563288 Publication Model: Electronic Cited Medium: Internet ISSN: 2045-2322 (Electronic) Linking ISSN: 20452322 NLM ISO Abbreviation: Sci Rep Subsets: MEDLINE; PubMed not MEDLINE |
| Imprint Name(s): | Original Publication: London : Nature Publishing Group, copyright 2011- |
| Περίληψη: | Earth System Models (ESM) are our main tool for projecting the impacts of climate change. However, running these models at sufficient resolution for local-scale risk-assessments is not computationally feasible. Deep learning-based super-resolution models offer a promising solution to downscale ESM outputs to higher resolutions by learning from data. Yet, due to regional variations in climatic processes, these models typically require retraining for each geographical area-demanding high-resolution observational data, which is unevenly available across the globe. This highlights the need to assess how well these models generalize across geographic regions. To address this, we introduce RainShift, a dataset and benchmark for evaluating downscaling under geographic distribution shifts. We evaluate state-of-the-art downscaling approaches including GANs and diffusion models in generalizing across data gaps between the Global North and Global South. Our findings reveal substantial performance drops in out-of-distribution regions, depending on model and geographic area. While expanding the training domain generally improves generalization, it is insufficient to overcome shifts between geographically distinct regions. We show that addressing these shifts through, for example, domain adaptation can improve spatial generalization. Our work advances the global applicability of downscaling methods and represents a step toward reducing inequities in access to high-resolution climate information. (© 2026. The Author(s).) |
| Competing Interests: | Declarations. Competing interests: The authors declare no competing interests. |
| References: | Sci Rep. 2022 Dec 16;12(1):21776. (PMID: 36526725) Sci Rep. 2017 Dec 21;7(1):17966. (PMID: 29269737) Nature. 2023 Jul;619(7969):305-310. (PMID: 37380773) Nature. 2023 Jul;619(7970):526-532. (PMID: 37407824) Nature. 2025 Jan;637(8044):84-90. (PMID: 39633054) J Adv Model Earth Syst. 2022 Oct;14(10):e2022MS003120. (PMID: 36590321) |
| Entry Date(s): | Date Created: 20260127 Latest Revision: 20260131 |
| Update Code: | 20260131 |
| PubMed Central ID: | PMC12852169 |
| DOI: | 10.1038/s41598-025-34557-4 |
| PMID: | 41593162 |
| Βάση Δεδομένων: | MEDLINE |
καταχωρήστε σχόλιο πρώτοι!