Machine Learning (ML)-Physics Fusion Model Outperforms Both Physics-Only and ML-Only Models in Typhoon Predictions
In one sentence. Hybrid FuXi-SHTM typhoon model cuts 2024 western Pacific intensity errors by 60% at 72 h versus FuXi and improves track errors over ECMWF HRES.
Abstract
Data-driven machine learning (ML) models, such as FuXi, exhibit notable limitations in forecasting typhoon intensity and structure. This study presents a comprehensive evaluation of FuXi-SHTM, a hybrid ML-physics model, using all 2024 western North Pacific typhoon cases. The FuXi-SHTM hybrid demonstrates clear improvements in both track and intensity forecasts compared to the standalone SHTM, FuXi, and ECMWF HRES models. Compared to FuXi alone, FuXi-SHTM reduces typhoon track forecast errors by 16.5% and 5.2% at lead times of 72 h and 120 h, respectively, and reduces intensity forecast errors by 59.7% and 47.6%. Furthermore, FuXi-SHTM simulates cloud structures more realistically compared to SHTM, and achieves superior representation of the 10-m wind fields in both intensity and spatial structure compared to FuXi and SHTM. Increasing the resolution of FuXi-SHTM from 9 km to 3 km further enhances intensity forecasts, highlighting the critical role of the resolution of the physical model in advancing hybrid forecasting capabilities.
Labels
- Topic
- Extremes & hazards, Global medium-range NWP, Hybrid physics-ML & parameterization
- Type
- New model or method
- Method
- Transformer / ViT, Physics-informed / physics-constrained
- Time scale
- 6 h–3 days, 3–15 days
- Domain
- Regional
- Data
- ERA5 / reanalysis, NWP analysis or forecasts (IFS, GFS, ...)
- Status
- Research
- Identifiers
- arXiv:2504.20852 OpenAlex:W4417172181
- Added
- 2026-10-22
Labels and summary are generated by Claude and reviewed weekly. Wrong? Edit this entry.