paper
2023
Post-processing & calibration
Evaluation or intercomparison of existing models
Classical ML (trees, GP, linear, kernels)3–15 daysECMWFUniversity of Oxford Journal
2023-01-12
051
Jan 2023
Zied Ben Bouallègue, Fenwick Cooper, Matthew Chantry et al.
Monthly Weather Review
ECMWF, University of Oxford
Linear regression, random forests and neural networks correct ECMWF IFS 2 m temperature and 10 m wind errors, improving RMSE by 10–15% with uncertainty estimates.
Statistical Modeling of 2-m Temperature and 10-m Wind Speed Forecast Errors
In one sentence. Linear regression, random forests and neural networks correct ECMWF IFS 2 m temperature and 10 m wind errors, improving RMSE by 10–15% with uncertainty estimates.
Labels
- Topic
- Post-processing & calibration
- Type
- Evaluation or intercomparison of existing models
- Method
- Classical ML (trees, GP, linear, kernels)
- Time scale
- 3–15 days
- Domain
- Global
- Data
- NWP analysis or forecasts (IFS, GFS, ...), Station / in-situ
- Status
- Research
- Identifiers
- DOI:10.1175/mwr-d-22-0107.1 OpenAlex:W4315706382
- Added
- 2026-10-22
Labels and summary are generated by Claude and reviewed weekly. Wrong? Edit this entry.