OneForecast: A Universal Framework for Global and Regional Weather Forecasting

Yuan Gao, Hao Wu, Ruiqi Shu, huanshuo dong, Fan Xu, Rui Ray Chen, Yibo Yan, Qingsong Wen, Xuming Hu, Kun Wang, Jiahao Wu, Li Qing, Hui Xiong, Xiaomeng Huang

Abstract: Accurate weather forecasts are important for disaster prevention, agricultural planning, etc. Traditional numerical weather prediction (NWP) methods offer physically interpretable high-accuracy predictions but are computationally expensive and fail to fully leverage rapidly growing historical data. In recent years, deep learning models have made significant progress in weather forecasting, but challenges remain, such as balancing global and regional high-resolution forecasts, excessive smoothing in extreme event predictions, and insufficient dynamic system modeling. To address these issues, this paper proposes a global-regional nested weather forecasting framework (OneForecast) based on graph neural networks. By combining a dynamic system perspective with multi-grid theory, we construct a multi-scale graph structure and densify the target region to capture local high-frequency features. We introduce an adaptive messaging mechanism, using dynamic gating units to deeply integrate node and edge features for more accurate extreme event forecasting. For high-resolution regional forecasts, we propose a neural nested grid method to mitigate boundary information loss. Experimental results show that OneForecast performs excellently across global to regional scales and short-term to long-term forecasts, especially in extreme event predictions. Codes link: \url{https://github.com/YuanGao-YG/OneForecast}.

Lay Summary:

Accurate weather forecasts are crucial for protecting communities from disasters, planning agriculture, and managing daily life. Traditional methods of weather prediction utilize well-understood physical principles but require substantial computing power and don't fully leverage the vast amounts of historical weather data now available. Recently, artificial intelligence (AI) has shown promise in improving weather forecasts, but existing AI models still struggle with precisely forecasting extreme weather events and balancing global coverage with local detail.

To tackle these challenges, we develop OneForecast, an AI-based weather prediction system that effectively combines large-scale global forecasting with detailed local predictions.

Our approach enables seamless weather forecasting and more accurate prediction of extreme events. In summary, our approach fosters the development of next-generation data-driven weather forecasting systems.

Application-Driven Machine Learning: This submission is on Application-Driven Machine Learning.

Link To Code: https://github.com/YuanGao-YG/OneForecast

Primary Area: Applications->Chemistry, Physics, and Earth Sciences

Keywords: AI for Science, AI for Earth, Weather Forecasting, Earth System Science

Submission Number: 1369