Northern hemisphere midlatitude cyclone intensity biases in machine learning weather prediction models

Dacre, H.F. ORCID: https://orcid.org/0000-0003-4328-9126, Charlton-Perez, A.J. ORCID: https://orcid.org/0000-0001-8179-6220, Driscoll, S., Gray, S.L. ORCID: https://orcid.org/0000-0001-8658-362X, Harvey, B. ORCID: https://orcid.org/0000-0002-6510-8181, Harvey, N.J. ORCID: https://orcid.org/0000-0003-0973-5794, Hodges, K.I. ORCID: https://orcid.org/0000-0003-0894-229X, Hunt, K.M.R. ORCID: https://orcid.org/0000-0003-1480-3755 and Volonté, A. ORCID: https://orcid.org/0000-0003-0278-952X (2025) Northern hemisphere midlatitude cyclone intensity biases in machine learning weather prediction models. Bulletin of the American Meteorological Society. ISSN 1520-0477 doi: 10.1175/BAMS-D-25-0129.1 (In Press)

Abstract/Summary

Forecasting the location and intensity of strong winds associated with midlatitude cyclones is important as they can have significant safety, economic, and environmental impacts. In this study we use a feature-based evaluation method to assess the performance of both numerical weather prediction and machine learning weather prediction (MLWP) models in forecasting midlatitude cyclone winds. By tracking over 1000 cyclones across the Northern Hemisphere from 1 October 2023 to 31 March 2024 in 7 MLWP models, we systematically compare model performance. Our results show that MLWP models predict midlatitude cyclone tracks with accuracy comparable to the ECMWF IFS-forecast out to 10 days. However, MLWP models exhibit a persistent intensity bias, underestimating cyclone minimum sea level pressure by more than 5~hPa at 10-day forecast lead times, whereas the ECMWF IFS-forecast has no bias. Additionally, all MLWP models produce weaker than observed peak 10-m winds, even at short lead times. In contrast, the ECMWF IFS forecast exhibits no bias in 10-m wind speed. These differences highlight the limitations of current MLWP models in capturing important high-impact weather features like peak wind speeds.

Item Type	Article
URI	https://centaur.reading.ac.uk/id/eprint/127752
Identification Number/DOI	10.1175/BAMS-D-25-0129.1
Refereed	Yes
Divisions	Science > School of Mathematical, Physical and Computational Sciences > Department of Meteorology
Publisher	American Meteorological Society
Download/View statistics	View download statistics for this item