Preprint · 2026 · Open access
Do AI weather models forecast extremes better than GFS? A station-based verification over China with a common initialization source, 2022-2025
Research Square · Published 6 October 2026 · DOI 10.21203/rs.3.rs-11259821/v1
In brief
“Do AI weather models forecast extremes better than GFS? A station-based verification over China with a common initialization source, 2022-2025” is a 2026 preprint by Yi Zou, published in Research Square. OpenAlex counts 0 citations as of 20 November 2026. It is open access under a CC BY licence: the full text is free to read at Research Square.
- Citations
- 0
- References
- 23
- Authors
- 1
Abstract
Whether artificial intelligence (AI) weather models are more informative for extremes depends on verification and initialization. We compare GraphCast Operational, Pangu-Weather and FourCastNet v2-small with the operational Global Forecast System (GFS) at 334 surface stations in China for 00 UTC initializations from January 2022 to August 2025. The AI runs share the GFS F000 input source, controlling initialization source while retaining model-specific training and preprocessing. Potential continuous ranked probability scores (PCRPS) and threshold-weighted versions assess information after in-sample isotonic distributional regression (IDR). Independent-period IDR and Gaussian ensemble model output statistics (EMOS) assess achievable post-processing skill. The temperature comparison favours the AI models at the reported daily leads, with exceptions at six-hourly leads. GraphCast's heavy-precipitation advantage weakens out of sample, where tail-skill intervals include zero beyond day 1. Temperature rankings differ between verification settings, skill varies across climate groups, and initialization from the Integrated Forecasting System (IFS) substantially changes short-range gains. Comparisons of extremes should distinguish potential information, achievable skill and initialization sensitivity.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Zou, Y. (2026). Do AI weather models forecast extremes better than GFS? A station-based verification over China with a common initialization source, 2022-2025. Research Square. https://doi.org/10.21203/rs.3.rs-11259821/v1
MLA 9
Zou, Yi. “Do AI weather models forecast extremes better than GFS? A station-based verification over China with a common initialization source, 2022-2025.” Research Square, 2026. https://doi.org/10.21203/rs.3.rs-11259821/v1.
Chicago
Zou, Yi. 2026. “Do AI weather models forecast extremes better than GFS? A station-based verification over China with a common initialization source, 2022-2025.” Research Square. https://doi.org/10.21203/rs.3.rs-11259821/v1.
Harvard
Zou, Y. (2026) ‘Do AI weather models forecast extremes better than GFS? A station-based verification over China with a common initialization source, 2022-2025’, Research Square. doi:10.21203/rs.3.rs-11259821/v1.
IEEE
Y. Zou, “Do AI weather models forecast extremes better than GFS? A station-based verification over China with a common initialization source, 2022-2025,” Research Square, 2026, doi: 10.21203/rs.3.rs-11259821/v1.
BibTeX
@misc{zou2026weather,
title = {{Do AI weather models forecast extremes better than GFS? A station-based verification over China with a common initialization source, 2022-2025}},
author = {Zou, Yi},
howpublished = {Research Square},
year = {2026},
doi = {10.21203/rs.3.rs-11259821/v1},
url = {https://doi.org/10.21203/rs.3.rs-11259821/v1}
}
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
- extreme weather prediction
- Pangu-Weather
- FourCastNet
- Global Forecast System
- continuous ranked probability score
- ensemble model output statistics
- temperature extremes
- heavy rainfall
Work with this paper
Most-cited papers in Earth and Planetary Sciences
- The ERA5 global reanalysisHans Hersbach, Bill Bell, Paul Berrisford and 40 more · Quarterly Journal of the Royal Meteorological Society · 2020 · 35,424 citations
- The NCEP/NCAR 40-Year Reanalysis ProjectEugenia Kalnay, Masao Kanamitsu, Robert E. Kistler and 19 more · Bulletin of the American Meteorological Society · 1996 · 28,486 citations
- Chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processesShen‐Su Sun, W. F. McDonough · Geological Society London Special Publications · 1989 · 26,603 citations
- The ERA‐Interim reanalysis: configuration and performance of the data assimilation systemDick P. Dee, Sakari M. Uppala, A. J. Simmons and 33 more · Quarterly Journal of the Royal Meteorological Society · 2011 · 25,245 citations
- PAST: PALEONTOLOGICAL STATISTICAL SOFTWARE PACKAGE FOR EDUCATION AND DATA ANALYSISØyvind Hammer, David A. T. Harper, Paul D. Ryan · 2001 · 17,871 citations
Questions about this paper
How do I cite this paper in APA style?
Zou, Y. (2026). Do AI weather models forecast extremes better than GFS? A station-based verification over China with a common initialization source, 2022-2025. Research Square. https://doi.org/10.21203/rs.3.rs-11259821/v1
How many times has this paper been cited?
0 times, according to OpenAlex on 20 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 Research Square: sign in and open the PDF from this page.
Who wrote this paper?
Yi Zou.
What is the DOI of this paper?
10.21203/rs.3.rs-11259821/v1, permanently at https://doi.org/10.21203/rs.3.rs-11259821/v1. Sign in and open it from this page.
Bibliographic data from OpenAlex (CC0), last checked 20 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.