paper
2025
Global medium-range NWP
Benchmarks, datasets & evaluation Evaluation or intercomparison of existing models
Graph neural network6 h–3 days3–15 daysColorado State University (incl. CIRA)NOAA (incl. GFDL, PSL, GSL, NCEP/EMC) Journal
★ Highlight 2025-03-03
059
Mar 2025
Jacob T Radford, Imme Ebert-Uphoff, Jebb Q. Stewart
Weather and Forecasting
Colorado State University (incl. CIRA), NOAA (incl. GFDL, PSL, GSL, NCEP/EMC)
Verification over CONUS finds GraphCast precipitation beats GFS and IFS on RMSE and equitable metrics, but NWP better captures precipitation intensity distributions and spatial placement.
1
Global medium-range NWP★ Highlight
A Comparison of AI Weather Prediction and Numerical Weather Prediction Models for 1–7-Day Precipitation Forecasts
Why it matters. Useful CONUS verification of GraphCast precipitation versus GFS and IFS, showing AI skill on RMSE but weaker intensity and placement.
In one sentence. Verification over CONUS finds GraphCast precipitation beats GFS and IFS on RMSE and equitable metrics, but NWP better captures precipitation intensity distributions and spatial placement.
Labels
- Topic
- Global medium-range NWP, Benchmarks, datasets & evaluation
- Type
- Evaluation or intercomparison of existing models
- Method
- Graph neural network
- Time scale
- 6 h–3 days, 3–15 days
- Domain
- Regional
- Data
- ERA5 / reanalysis, NWP analysis or forecasts (IFS, GFS, ...), Radar
- Status
- Research
- Identifiers
- DOI:10.1175/waf-d-24-0081.1 OpenAlex:W4408112191
- Added
- 2026-10-22
Labels and summary are generated by Claude and reviewed weekly. Wrong? Edit this entry.