Articles | Volume 18, issue 2
https://doi.org/10.5194/essd-18-1379-2026
https://doi.org/10.5194/essd-18-1379-2026
Data description article
 | 
24 Feb 2026
Data description article |  | 24 Feb 2026

GlobalGeoTree: a multi-granular vision-language dataset for global tree species classification

Yang Mu, Zhitong Xiong, Yi Wang, Muhammad Shahzad, Franz Essl, Holger Kreft, Mark van Kleunen, and Xiao Xiang Zhu

Data sets

GlobalGeoTree: A Multi-Granular Vision-Language Dataset for Global Tree Species Classification Y. Mu et al. https://doi.org/10.15468/dd.9qxqyy

Model code and software

MUYang99/GlobalGeoTree: GlobalGeoTree v1.0.0 (v1.0.0) Y. Mu et al. https://doi.org/10.5281/zenodo.18619541

Download
Short summary
To better protect our planet's forests, we need to know what trees are where. We created GlobalGeoTree, a massive public dataset linking 6.3 million tree locations worldwide with satellite data. This dataset helps computers learn to identify tree species from space, supporting biodiversity monitoring and climate action. Our baseline model shows this is a promising path to understanding global forests.
Share
Altmetrics
Final-revised paper
Preprint