paper
2023
Post-processing & calibration
Explainability, uncertainty & physics New model or method
Classical ML (trees, GP, linear, kernels)Bureau of Meteorology (Australia) Journal
2023-02-10
026
Feb 2023
Marc D. Mallet, Simon Peter Alexander, Alain Protat et al.
Artificial Intelligence for the Earth Systems
Bureau of Meteorology (Australia)
XGBoost and random forest models trained on 25 years of Southern Ocean surface observations cut ERA5 shortwave mean bias to near zero and reduce RMSE by about 25%.
Reducing Southern Ocean Shortwave Radiation Errors in the ERA5 Reanalysis with Machine Learning and 25 Years of Surface Observations
In one sentence. XGBoost and random forest models trained on 25 years of Southern Ocean surface observations cut ERA5 shortwave mean bias to near zero and reduce RMSE by about 25%.
Labels
- Topic
- Post-processing & calibration, Explainability, uncertainty & physics
- Type
- New model or method
- Method
- Classical ML (trees, GP, linear, kernels)
- Domain
- Regional
- Data
- ERA5 / reanalysis, Station / in-situ
- Status
- Research
- Identifiers
- DOI:10.1175/aies-d-22-0044.1 OpenAlex:W4319963651
- Added
- 2026-10-22
Labels and summary are generated by Claude and reviewed weekly. Wrong? Edit this entry.