Search papers, labs, and topics across Lattice.
FuXiWeather2 is introduced as an end-to-end neural framework for global weather forecasting that unifies data assimilation and numerical modeling. The model is trained using a combination of real-world observations and reanalysis data, along with a recursive unrolling training method to improve analysis generation and handle distribution shifts. Results show that FuXiWeather2 generates high-resolution global analysis fields and 10-day forecasts within minutes, outperforming NCEP-GFS, ERA5, and ECMWF-HRES in various metrics, including typhoon track prediction.
A new AI weather model beats the gold-standard ECMWF-HRES system in 91% of metrics, while also providing faster analysis and forecasts.
Numerical weather prediction has long been constrained by the computational bottlenecks inherent in data assimilation and numerical modeling. While machine learning has accelerated forecasting, existing models largely serve as "emulators of reanalysis products," thereby retaining their systematic biases and operational latencies. Here, we present FuXiWeather2, a unified end-to-end neural framework for assimilation and forecasting. We align training objectives directly with a combination of real-world observations and reanalysis data, enabling the framework to effectively rectify inherent errors within reanalysis products. To address the distribution shift between NWP-derived background inputs during training and self-generated backgrounds during deployment, we introduce a recursive unrolling training method to enhance the precision and stability of analysis generation. Furthermore, our model is trained on a hybrid dataset of raw and simulated observations to mitigate the impact of observational distribution inconsistency. FuXiWeather2 generates high-resolution ($0.25^{\circ}$) global analysis fields and 10-day forecasts within minutes. The analysis fields surpass the NCEP-GFS across most variables and demonstrate superior accuracy over both ERA5 and the ECMWF-HRES system in lower-tropospheric and surface variables. These high-quality analysis fields drive deterministic forecasts that exceed the skill of the HRES system in 91\% of evaluated metrics. Additionally, its outstanding performance in typhoon track prediction underscores its practical value for rapid response to extreme weather events. The FuXiWeather2 analysis dataset is available at https://doi.org/10.5281/zenodo.18872728.
