Deep learning for post-processing ensemble weather forecasts
Problem
Ensemble weather forecasts are computationally expensive and their raw predictions need improvement, especially for extreme weather events.
Why it matters
Better extreme weather prediction saves lives and reduces economic damage from storms and heatwaves.
What the AI did
Deep neural networks post-process a subset of ensemble trajectories, capturing non-linear relationships missed by numerical models.
Results
Relative improvement in ensemble forecast skill (CRPS) of over 14% on global data; larger improvement for extreme weather events in select case studies.
Impact
Demonstrated measurable improvement over existing post-processing methods on real forecast data.
Record
| Venue | Philosophical Transactions of the Royal Society A Mathematical Physical and Engineering Sciences |
|---|---|
| Published | 2021 |
| Work type | article |
| Authors | Peter Grönquist, Chengyuan Yao, Tal Ben‐Nun, Nikoli Dryden, Peter Dominik Dueben, Shigang Li, Torsten Hoefler |
| Citations | 196 (OpenAlex, at import time) |
| References | 55 |
| Model or method | deep neural networks (not further specified) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1098/rsta.2020.0092 |
| Open access | green |
| Keywords | deep learning, weather forecasting, ensemble post-processing, CRPS, extreme weather |
Importance rating
significant — thousands to millions affected
Improving extreme-weather ensemble forecasts plausibly protects lives and reduces damage, and >14% CRPS gain is meaningful, but abstract quantifies no human or economic scale.
More in Climate & Environment
| # | Paper | Year | Citations | Importance |
|---|---|---|---|---|
| 1 | Learning skillful medium-range global weather forecasting | 2023 | 1,416 |
8
|
| 2 | Accurate medium-range global weather forecasting with 3D neural networks | 2023 | 1,631 |
8
|
| 3 | FuXi: a cascade machine learning forecasting system for 15-day global weather forecast | 2023 | 355 |
8
|
| 4 | Skilful nowcasting of extreme precipitation with NowcastNet | 2023 | 451 |
8
|
| 5 | Forecasting Renewable Energy Generation with Machine Learning and Deep Learning: Current Advances and Future Prospects | 2023 | 378 |
8
|