@article{59f76545d25c4774aebbfb7ded0c2c91,
title = "Process-based machine learning observationally constrains future regional warming projections",
abstract = "We present the results of a novel process-based machine learning method to constrain climate model uncertainty in future regional temperature projections. Ridge-ERA5 - a ridge regression model - learns coefficients to represent observed relationships between daily near-surface temperature anomalies and predictor variables from ERA5 reanalysis in Northern Hemisphere land regions. Combining the historically constrained Ridge-ERA5 coefficients with inputs from CMIP6 future projections enables a derivation of observational constraints on regional warming. Although the multi-model mean falls within the constrained range of temperatures in all tested regions, a subset of models which predict the greatest degree of warming tend to be excluded and decomposition of the constraint into predictor variable contributions suggests error-cancellation of feedbacks in some models and regions.",
author = "Sophie Wilkinson and Peer Nowack and Manoj Joshi",
note = "Data Availability Statement: All reanalysis and CMIP6 data used in this study are publicly available. A specific list of data sets and variables accessed, data pre-processing steps, and Python code for analysis are archived in Zenodo and available at: https://doi.org/10.5281/zenodo.15115895 (Wilkinson, 2025). Acknowledgments: S.W. was supported by a University of East Anglia Faculty of Science PhD studentship. P.N. and M.J. were supported by the UK Natural Environment Research Council (Grant NE/V012045/1). We acknowledge the WCRP, which, through its Working Group on Coupled Modeling, coordinated and promoted CMIP6. We thank the climate-modeling groups for producing and making available their model output, the Earth System Grid Federation (ESGF) for archiving the data and providing access and the multiple funding agencies that support CMIP6 and ESGF. We further used the JASMIN postprocessing system (Lawrence et al., 2013) operated by the Science and Technology Facilities Council on behalf of the UK Natural Environment Research Council. The analysis presented in this paper was mostly carried out on the High Performance Computing Cluster supported by the Research and Specialist Computing Support service at the University of East Anglia.",
year = "2025",
month = jun,
doi = "10.1029/2025JH000698",
language = "English",
volume = "2",
journal = "Journal of Geophysical Research: Machine Learning and Computation",
issn = "2993-5210",
publisher = "Wiley",
number = "2",
}
