What the study found
Post-processing generally improved both probabilistic calibration and point forecast accuracy for wind-speed ensemble forecasts. The study also found that spatial resolution was more important than ensemble size in these forecasts.
Why the authors say this matters
The authors conclude that forecast skill can be changed substantially by statistical calibration, which reduces differences between forecast configurations. They also suggest that adding low-resolution forecasts to a sufficiently large high-resolution ensemble does not necessarily improve skill, while adding high-resolution members to low-resolution forecasts can provide clear gains.
What the researchers tested
The researchers compared raw and post-processed medium- and extended-range wind-speed ensemble forecasts from the European Centre for Medium-Range Weather Forecasts at 9 km and 36 km horizontal resolutions. They used the ensemble model output statistic approach, a statistical calibration method, with three spatial training data selection techniques, and tested multiple mixtures of high- and low-resolution ensemble members.
What worked and what didn't
Across configurations, post-processed forecasts generally outperformed raw ensemble predictions in both calibration and point forecast accuracy. Post-processing also reduced differences between the various ensemble setups. Adding low-resolution forecasts to a sufficiently large high-resolution ensemble did not necessarily improve forecast skill, while adding high-resolution members to low-resolution forecasts showed the clearest gains, especially when more high-resolution members were included.
What to keep in mind
The abstract does not describe detailed limitations beyond the specific forecast systems, resolutions, and time ranges studied. The findings are limited to the wind-speed ensemble forecasts and calibration setups examined here.
Key points
- Post-processing generally improved calibration and point forecast accuracy for wind-speed ensemble forecasts.
- Spatial resolution was found to be more important than ensemble size.
- Adding low-resolution forecasts to a sufficiently large high-resolution ensemble did not necessarily improve skill.
- Adding high-resolution members to low-resolution ensembles produced the clearest gains.
- Post-processing reduced differences between the forecast configurations.
Disclosure
- Research title:
- Post-processing improved wind-speed ensemble forecasts
- Authors:
- Sándor Baran, M Lakatos
- Institutions:
- University of Debrecen
- Publication date:
- 2026-04-20
- DOI:
- 10.1002/qj.70201
- OpenAlex record:
- View
Get the weekly research newsletter
Stay current with peer-reviewed research without reading academic papers — one filtered digest, every Friday.