Preprint · 2026 · Open access

Beyond the record: evaluating probabilistic AI weather forecasts deep in the tail

Bastiaan Van Velthoven, Cas Decancq, Diego G. Miralles, Stijn Luca, Thomas Mortier

arXiv (Cornell University) · Published 18 October 2026 · DOI 10.48550/arxiv.2610.25545

Open PDFView at the publisherCite this paper

In brief

“Beyond the record: evaluating probabilistic AI weather forecasts deep in the tail” is a 2026 preprint by Bastiaan Van Velthoven and 4 co-authors, published in arXiv (Cornell University). OpenAlex counts 0 citations as of 18 November 2026. It is open access under a CC BY licence: the full text is free to read at arXiv (Cornell University).

Citations
0
References
58
Authors
5

Abstract

Can probabilistic artificial intelligence (AI) weather forecasts retain skill beyond their training domain? We compare four state-of-the-art AI models (ArchesWeatherGen, U-Cast, WeatherNext2 and GenCast-mini) with the physics-based IFS-ENS over land in 2022, at 1.5° resolution and lead times of 2--10 days. Monthly ERA5 records from 1979--2021 define common historical boundaries, and we characterize new records set in 2022 by their depth, i.e., the exceedance margin relative to local historical variability. Two proper scores, record-threshold weighted CRPS and Brier score, assess the magnitude and probability of record exceedances across all eligible cases, retaining both missed extremes and false alarms. Verified against ERA5, AI models outperform IFS-ENS for heat and cold records at almost every lead and depth. For heat, the advantage strengthens with depth, partly because IFS-ENS over-forecasts deep records, and largely persists for deeper records under HRES-fc0 verification. For wind, gains erode for deeper records and longer leads. Better tail scores can nevertheless coexist with aggregate probability deficits or surpluses, and some improvements arise chiefly on non-exceedances. Increasing underprediction on progressively more extreme observed subsets can also arise through outcome selection, even for an ideal conditional-mean forecast. These findings establish that native probabilistic AI ensembles retain, and for heat increase, their advantage over IFS-ENS as thresholds move beyond local records, although all advantages narrow with lead time. They caution against generalizing the extrapolation failure reported for deterministic AI forecasts to probabilistic systems, and motivate common verification conventions that separate comparative tail skill, performance on realized extremes, and probabilistic calibration.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Van Velthoven, B., Decancq, C., Miralles, D. G., Luca, S., & Mortier, T. (2026). Beyond the record: evaluating probabilistic AI weather forecasts deep in the tail. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2610.25545

MLA 9

Van Velthoven, Bastiaan, et al. “Beyond the record: evaluating probabilistic AI weather forecasts deep in the tail.” arXiv (Cornell University), 2026. https://doi.org/10.48550/arxiv.2610.25545.

Chicago

Van Velthoven, Bastiaan, Cas Decancq, Diego G. Miralles, Stijn Luca, and Thomas Mortier. 2026. “Beyond the record: evaluating probabilistic AI weather forecasts deep in the tail.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2610.25545.

Harvard

Van Velthoven, B. et al. (2026) ‘Beyond the record: evaluating probabilistic AI weather forecasts deep in the tail’, arXiv (Cornell University). doi:10.48550/arxiv.2610.25545.

IEEE

B. Van Velthoven, C. Decancq, D. G. Miralles, S. Luca, and T. Mortier, “Beyond the record: evaluating probabilistic AI weather forecasts deep in the tail,” arXiv (Cornell University), 2026, doi: 10.48550/arxiv.2610.25545.

BibTeX

@misc{vanvelthoven2026beyond,
  title     = {{Beyond the record: evaluating probabilistic AI weather forecasts deep in the tail}},
  author    = {Van Velthoven, Bastiaan and Decancq, Cas and Miralles, Diego G. and Luca, Stijn and Mortier, Thomas},
  howpublished = {arXiv (Cornell University)},
  year      = {2026},
  doi       = {10.48550/arxiv.2610.25545},
  url       = {https://doi.org/10.48550/arxiv.2610.25545}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Topic and keywords

Physical Sciences › Earth and Planetary Sciences › Atmospheric Science › Meteorological Phenomena and Simulations

  • probabilistic weather forecasting
  • proper scoring rules
  • complex regional pain syndrome
  • Brier score
  • ensemble forecasting
  • ERA5 reanalysis
  • extreme heat
  • cold extremes

Work with this paper

Most-cited papers in Earth and Planetary Sciences

All highly cited Earth and Planetary Sciences papers

Questions about this paper

How do I cite this paper in APA style?

Van Velthoven, B., Decancq, C., Miralles, D. G., Luca, S., & Mortier, T. (2026). Beyond the record: evaluating probabilistic AI weather forecasts deep in the tail. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2610.25545

How many times has this paper been cited?

0 times, according to OpenAlex on 18 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access at arXiv (Cornell University): sign in and open the PDF from this page.

Who wrote this paper?

Bastiaan Van Velthoven, Cas Decancq, Diego G. Miralles, Stijn Luca, Thomas Mortier.

What is the DOI of this paper?

10.48550/arxiv.2610.25545, permanently at https://doi.org/10.48550/arxiv.2610.25545. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 18 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.