paper
2021
Post-processing & calibration
Local & station forecasting New model or method
Transformer / ViTCNN / U-Net6 h–3 daysInstitute of Atmospheric Physics, Chinese Academy of Sciences Journal
2021-12-05
022
Dec 2021
Wenjia Kong, Haochen Li, Chen Yu et al.
Communications in Computational Physics
Institute of Atmospheric Physics, Chinese Academy of Sciences
DeepSTF combines CNN and attention encoder-decoder to post-process multi-site forecasts at 226 Beijing stations, improving on MOS benchmarks.
A Deep Spatio-Temporal Forecasting Model for Multi-Site Weather Prediction Post-Processing
In one sentence. DeepSTF combines CNN and attention encoder-decoder to post-process multi-site forecasts at 226 Beijing stations, improving on MOS benchmarks.
Labels
- Topic
- Post-processing & calibration, Local & station forecasting
- Type
- New model or method
- Method
- Transformer / ViT, CNN / U-Net
- Time scale
- 6 h–3 days
- Domain
- Local / station
- Data
- Station / in-situ, NWP analysis or forecasts (IFS, GFS, ...)
- Status
- Research
- Identifiers
- DOI:10.4208/cicp.oa-2020-0158 OpenAlex:W4200479603
- Added
- 2026-10-22
Labels and summary are generated by Claude and reviewed weekly. Wrong? Edit this entry.