FuXiWeather2: Learning accurate atmospheric state estimation for operational global weather forecasting

2026-03-16

AI Summary

This work proposes the first end-to-end neural network framework that jointly optimizes data assimilation and numerical weather prediction, addressing computational bottlenecks in traditional approaches and the systematic biases of existing machine learning models that merely emulate reanalysis data. By fusing real observations with reanalysis products, employing a recurrent unrolling training strategy to mitigate background field distribution shifts, and incorporating hybrid observations to enhance generalization, the model generates minute-scale analysis fields and 10-day forecasts on a global 0.25° grid. It outperforms NCEP-GFS across most variables and surpasses ECMWF-HRES on 91% of evaluation metrics, demonstrating particularly strong performance in near-surface variables and tropical cyclone track prediction.

Technology Category

• Natural Language Processing: Learning & Optimization for NLP
• Machine Learning: Evaluation and Analysis
• Search and Optimization: Learning to Search

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

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.

Problem

• numerical weather prediction
• data assimilation
• reanalysis bias
• computational bottleneck
• global weather forecasting

Innovation

• end-to-end neural framework
• recursive unrolling training
• data assimilation
• distribution shift mitigation
• hybrid observation dataset