Global medium-range NWP★ Highlight

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

Junchao Gong, Kaiyi Xu, Wangxu Wei, Siwei Tu, Jingyi Xu, Zili Liu, Hang Fan, Zhiwang Zhou, Tao Han, Yi Xiao, Xinyu Gu, Zhangrui Li et al.

arXiv preprint · 07 May 2026 · China Meteorological Administration (CMA), Columbia University (LEAP, Lamont-Doherty), Fudan University (FuXi), Institute of Atmospheric Physics, Chinese Academy of Sciences, Shanghai Artificial Intelligence Laboratory

Paper PDF

Why it matters. Grid-free observation-native world model reaching IFS-comparable surface skill; promising step toward end-to-end observation-driven forecasting, though early claims need validation.

In one sentence. Earth-o1 is a grid-free world model that learns atmospheric dynamics directly from ungridded observations, reaching surface forecast skill comparable to ECMWF IFS in hindcasts.

Abstract

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.

Labels

Labels and summary are generated by Claude and reviewed weekly. Wrong? Edit this entry.