Articles | Volume 18, issue 10
https://doi.org/10.5194/essd-18-7763-2026
https://doi.org/10.5194/essd-18-7763-2026
Data description article
 | 
28 Oct 2026
Data description article |  | 28 Oct 2026

BorFIT: a novel LiDAR-based training dataset for individual tree segmentation and species detection in northern boreal forests

Jacob Schladebach, Birgit Heim, Léa Enguehard, Mareike Wieczorek, Jakob Broers, Robert Jackisch, Josias Gloy, Kunyan Hao, James Tretton, Anna Gorshunova, and Stefan Kruse

Data sets

Reference dataset of individual trees from the Tundra-Taiga-Ecotone and Northern boreal forests BorFIT Stefan Kruse et al. https://doi.org/10.1594/PANGAEA.980505

Reference dataset of individual trees from the Tundra-Taiga-Ecotone and Northern boreal forests BorFIT Jacob Schladebach et al. https://doi.org/10.1594/PANGAEA.980505

Point clouds with ground point classification and individual tree segmentation of 89 northern boreal forest and tundra sites from UAV-based lidar surveys in Yakutia, Russia in 2021 Stefan Kruse et al. https://doi.org/10.1594/PANGAEA.980735

Point clouds with ground point classification and individual tree segmentation of 25 northern boreal forest and tundra sites from UAV-based lidar surveys in northwestern Canada in 2022 Stefan Kruse et al. https://doi.org/10.1594/PANGAEA.977771

Point clouds with ground point classification and individual tree segmentation of 47 northern boreal forest and tundra sites from UAV-based lidar surveys in the eastern part of Alaska in 2023 Stefan Kruse et al. https://doi.org/10.1594/PANGAEA.980485

Point clouds with ground point classification and individual tree segmentation of 28 northern boreal forest and tundra sites from UAV-based lidar surveys in western and central parts of Alaska in 2024 Stefan Kruse et al. https://doi.org/10.1594/PANGAEA.980757

Model code and software

BorFIT Jacob Schladebach https://github.com/StefanKruse/BorFIT/tree/main

Download
Short summary
BorFIT is a novel training dataset for light detection and ranging (LiDAR) point cloud segmentation and tree species detection in boreal forests. Comprising 384 plots across Siberia, Canada, and Alaska, it features 16 530 manually segmented trees of 12 species. BorFIT supports artificial intelligence (AI) applications for analyzing species distribution, stand structure, and boreal forest response to climate change.
Share
Altmetrics
Final-revised paper
Preprint