Reconstruction of tropical cyclone boundary layer wind field using physics-informed machine learning
Problem
Reconstructing high-fidelity 3D boundary layer wind fields of tropical cyclones from limited observations is difficult.
Why it matters
Accurate cyclone wind-field reconstruction underpins hazard assessment for populated coastal regions.
What the AI did
A physics-informed machine learning model embeds the governing wind-field equations with spatially varying eddy diffusivity in its loss function and assimilates satellite, dropsonde and radar data to reconstruct two cyclone events.
Results
Accuracy improves with more real data and spatially varying eddy diffusivity; compared against observations and a classic linear model, but no numeric metrics stated in abstract
Impact
Applied to two tropical cyclone events in different oceanic basins using multi-source observational data.
Record
| Venue | Physics of Fluids |
|---|---|
| Published | 2024 |
| Work type | article |
| Authors | Feng Hu, Q.S. Li |
| Citations | 13 (OpenAlex, at import time) |
| References | 67 |
| Model or method | physics-informed machine learning model (no specific architecture named) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1063/5.0234728 |
| Open access | green |
| Keywords | physics-informed machine learning, tropical cyclones, boundary layer, wind field reconstruction, eddy diffusivity |
Importance rating
significant — thousands to millions affected
Reconstructing tropical cyclone wind fields supports hazard-relevant science; applied to two cyclone events with multi-source observations, but the abstract states no population or impact 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
|