arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2605.06337cs.CV
收藏

Earth-o1:一种无网格的观测本征大气世界模型

Earth-o1: A Grid-free Observation-native Atmospheric World Model

Shanghai Artificial Intelligence Laboratory(上海人工智能实验室);Department of Information Engineering(信息工程系);School of Electronic Information and Electrical Engineering(电子信息与电气工程学院);Department of Atmospheric and Oceanic Sciences(大气与海洋科学系);School of Information Science and Technology(信息科学与技术学院);State Key Laboratory of Earth System Numerical Modeling and Application, Institute of Atmospheric Physics(地球系统数值模拟与应用国家重点实验室,大气物理研究所);College of Computer Science and Artificial Intelligence(计算机科学与人工智能学院);Department of Earth and Environmental Engineering(地球与环境工程系);Chinese Academy of Meteorological Sciences(中国气象科学研究院);School of Atmospheric Sciences(大气科学学院)
Junchao Gong, Kaiyi Xu, Wangxu Wei, Siwei Tu, Jingyi Xu, Zili Liu, Hang Fan, Zhiwang Zhou, Tao Han, Yi Xiao, Xinyu Gu, Zhangrui Li, Wenlong Zhang, Hao Chen, Xia… 展开作者

Junchao Gong, Kaiyi Xu, Wangxu Wei, Siwei Tu, Jingyi Xu, Zili Liu, Hang Fan, Zhiwang Zhou, Tao Han, Yi Xiao, Xinyu Gu, Zhangrui Li, Wenlong Zhang, Hao Chen, Xiaokang Yang, Yaqiang Wang, Lijing Cheng, Pierre Gentine, Wanli Ouyang, Feng Zhang, Zhe-Min Tan, Bowen Zhou, Fenghua Ling, Ben Fei, Lei Bai

AI总结:

Earth-o1通过直接学习无网格观测数据中的连续三维物理演变,实现对大气状态的自主时空推进,无需传统数值求解器,其预测精度可与现有物理框架相媲美。

摘要:

尽管现代地球观测系统提供了前所未有的多模态数据量,但我们的大气动力学建模能力仍受限制。传统建模框架将异构测量强制放入预定义的空间网格中,限制了原始传感器数据的全面利用并造成严重的计算瓶颈。本文提出Earth-o1,一种观测本征的大气世界模型,克服了这些结构限制。该模型不依赖传统大气动力学建模系统或传统数据同化,而是直接从非网格化观测数据中学习地球系统的连续三维物理演变。通过将多样传感器输入整合到统一的无网格动态场中,模型能够自主推进大气状态在空间和时间上。我们证明,这种根本不同的范式能够实现直接、实时的预测和跨传感器推断,而无需显式的数值求解器。在回溯评估中,Earth-o1在表面预测技能上与运营的集成预报系统(IFS)相当。这些结果表明,连续、观测驱动的世界模型——一种全新的完全观测本征的地球物理模拟器——可以与现有物理框架相媲美,为地球的数字孪生提供可扩展的数据驱动基础。

英文摘要:

Despite the unprecedented volume of multimodal data provided by modern Earth observation systems, our ability to model atmospheric dynamics remains constrained. Traditional modeling frameworks force heterogeneous measurements into predefined spatial grids, inherently limiting the full exploitation of raw sensor data and creating severe computational bottlenecks. Here we present Earth-o1, an observation-native atmospheric world model that overcomes these structural limitations. Rather than relying on conventional atmospheric dynamical modeling systems or traditional data assimilation, Earth-o1 directly learns the continuous, three-dimensional physical evolution of the Earth system from ungridded observational data. By integrating diverse sensor inputs into a unified, grid-free dynamical field, the model autonomously advances the atmospheric state in space and time. We show that this fundamentally distinct paradigm enables direct, real-time forecasting and cross-sensor inference without the overhead of explicit numerical solvers. In hindcast evaluations, Earth-o1 achieves surface forecast skill comparable to the operational Integrated Forecasting System (IFS). These results establish that continuous, observation-driven world models -- a new class of fully observation-native geophysical simulators -- can match the fidelity of established physical frameworks, providing a scalable data-driven foundation for a digital twin of the Earth.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑