paper
2023
Downscaling & super-resolution
Post-processing & calibration New model or method
Transformer / ViT6 h–3 daysAlibaba (DAMO Academy)Fudan University (FuXi) Journal
2023-10-14
051
Oct 2023
Xiaohui Zhong, Du Fei, Lei Chen et al.
Quarterly Journal of the Royal Meteorological Society
Alibaba (DAMO Academy), Fudan University (FuXi)
SwinIR and Uformer transformers downscale and bias-correct ECMWF HRES 2 m temperature and 10 m wind to 0.01°, outperforming UNet and RCAN baselines.
Downscaling & super-resolution
Investigating transformer‐based models for spatial downscaling and correcting biases of near‐surface temperature and wind‐speed forecasts
In one sentence. SwinIR and Uformer transformers downscale and bias-correct ECMWF HRES 2 m temperature and 10 m wind to 0.01°, outperforming UNet and RCAN baselines.
Labels
- Topic
- Downscaling & super-resolution, Post-processing & calibration
- Type
- New model or method
- Method
- Transformer / ViT
- Time scale
- 6 h–3 days
- Domain
- Regional
- Data
- NWP analysis or forecasts (IFS, GFS, ...)
- Status
- Research
- Identifiers
- DOI:10.1002/qj.4596 OpenAlex:W4387642293
- Added
- 2026-10-22
Labels and summary are generated by Claude and reviewed weekly. Wrong? Edit this entry.