Abstract
Statistical downscaling translates coarse-resolution climate model output into locally relevant information for climate services and impact assessment. Recent advances in artificial intelligence (AI) enable high-resolution, probabilistic, and computationally efficient approaches. This paper provides a perspective on the evolution from classical to AI-driven and hybrid downscaling approaches, assesses key challenges related to interpretability, uncertainty, data availability, and computational requirements, and outlines physically constrained and generative frameworks that support decision-making across sectors.
Similar content being viewed by others
AI for climate impacts: applications in flood risk
Artificial intelligence for modeling and understanding extreme weather and climate events
Early warning of complex climate risk with integrated artificial intelligence
Acknowledgements
This work was supported by the NERC–FAPESP–NSTC Land Use Change Investigation and Regional Climate (LIRIC) project [NE/Z504026/1], and by Climate Collaboratorium: Co-creation of Applied Theatre Decision Labs for exploring Climate Adaptation and Mitigation, part of the 2023 International Joint Initiative for Research in Climate Change Adaptation and Mitigation Competition, funded by the Economic and Social Research Council (ESRC)/UKRI [ES/Z000238/1]. Matías Ezequiel Olmo is funded by the AI4Science PN070500 fellowship within the “Generación D” initiative, Red.es, Ministerio para la Transformación Digital y de la Función Pública, for talent attraction (C005/24-ED CV1). Funded by the European Union NextGenerationEU funds, through PRTR. V.D.N. was supported by the German Federal Ministry of Education and Research (projects 01LP1903E and 01LP2324E) within the ClimXtreme network (FONA3). Konstantinos V. Varotsos is funded by the European Union’s Horizon 2020 research and innovation project MOIRAI (Grant Agreement. No. 101180994) This work is a contribution to the Med-CORDEX initiative (https://med-cordex.github.io/), and the Seasonal-to-decadal climate predictability in the Mediterranean: process understanding and services (MEDUSSE) COST Action (CA23108) It is also a part of the CORDEX‑WRF community within the WCRP Flagship Pilot Study URB‑RCC and the NCAR South American Affinity Group, and also contributes to the National Federated Compute Services NetworkPlus initiative, Enhancing HPC Adoption Through User‑Centred Design: A Roadmap for Inclusive Innovation in Environment, Health, and the Built Environment. It also informs the International Association of Hydrological Sciences (IAHS) Water Solutions Decade, supporting efforts to decolonise hydrology and promote diversity, inclusiveness, and equality, while contributing to Theme C (Compound Risks) and Theme D (Machine Learning and AI) of the UNESCO EURO‑FRIEND Water Project‑3 on Large‑Scale Variations in Hydrological Characteristics.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
41612_2026_1424_MOESM1_ESM (download PDF )
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Reprints and permissions
About this article
Cite this article
Chun, K.P., Aragão, L., Olmo, M.E. et al. New horizons in statistical downscaling and AI approaches for sustainable km-scale climate simulations.
npj Clim Atmos Sci (2026). https://doi.org/10.1038/s41612-026-01424-6
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41612-026-01424-6
Source: Resources - nature.com
