Articles | Volume 17, issue 2
https://doi.org/10.5194/essd-17-351-2025
© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/essd-17-351-2025
© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
A Sentinel-2 machine learning dataset for tree species classification in Germany
Maximilian Freudenberg
CORRESPONDING AUTHOR
Forest Inventory and Remote Sensing, University of Göttingen, Göttingen, Germany
Neural Data Science Group, University of Göttingen, Göttingen, Germany
Sebastian Schnell
Thünen Institute of Forest Ecosystems, Eberswalde, Germany
Faculty of Resource Management, University of Applied Sciences and Arts (HAWK), Göttingen, Germany
Viewed
Total article views: 6,870 (including HTML, PDF, and XML)
Cumulative views and downloads
(calculated since 04 Jul 2024)
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 4,825 | 1,818 | 227 | 6,870 | 148 | 212 |
- HTML: 4,825
- PDF: 1,818
- XML: 227
- Total: 6,870
- BibTeX: 148
- EndNote: 212
Total article views: 4,697 (including HTML, PDF, and XML)
Cumulative views and downloads
(calculated since 03 Feb 2025)
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 3,625 | 935 | 137 | 4,697 | 106 | 137 |
- HTML: 3,625
- PDF: 935
- XML: 137
- Total: 4,697
- BibTeX: 106
- EndNote: 137
Total article views: 2,173 (including HTML, PDF, and XML)
Cumulative views and downloads
(calculated since 04 Jul 2024)
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 1,200 | 883 | 90 | 2,173 | 42 | 75 |
- HTML: 1,200
- PDF: 883
- XML: 90
- Total: 2,173
- BibTeX: 42
- EndNote: 75
Viewed (geographical distribution)
Total article views: 6,870 (including HTML, PDF, and XML)
Thereof 6,690 with geography defined
and 180 with unknown origin.
Total article views: 4,697 (including HTML, PDF, and XML)
Thereof 4,637 with geography defined
and 60 with unknown origin.
Total article views: 2,173 (including HTML, PDF, and XML)
Thereof 2,053 with geography defined
and 120 with unknown origin.
| Country | # | Views | % |
|---|
| Country | # | Views | % |
|---|
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
1
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
1
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
1
Cited
15 citations as recorded by crossref.
- Implementing a growth pattern of climax community as post-processing filters core to improve tree species classification accuracy B. Zhang et al. https://doi.org/10.1080/17538947.2025.2498601
- Multi-Temporal Fusion of Sentinel-1 and Sentinel-2 Data for High-Accuracy Tree Species Identification in Subtropical Regions H. Li et al. https://doi.org/10.3390/rs18040592
- Integrating ChloroNet and XAI for accurate SPAD prediction in rice M. Abbasi et al. https://doi.org/10.1016/j.rsase.2026.101971
- Combining Sentinel-2 time-series and SITS-BERT for area-wide mapping of eight central European tree species J. Költzow et al. https://doi.org/10.1093/forestry/cpag042
- Geometry-Constrained Reference Sample Construction from Forest Inventory Compartments for Dominant Tree Species Mapping P. Zheng et al. https://doi.org/10.3390/rs18172915
- Multi-temporal Sentinel-2 images and LiDAR data fusion in dominant tree species classification D. Rim & I. Farah https://doi.org/10.1007/s41060-026-01218-2
- Identifying (less) vulnerable forest structures to storms in different conifer-dominated European forest landscapes by simple spaceborne proxies B. Garamszegi et al. https://doi.org/10.1139/cjfr-2025-0115
- Predicting rice crop height from field and Sentinel-2 data M. Abbasi et al. https://doi.org/10.1080/01431161.2025.2582213
- Large-scale mapping and uncertainty assessment of dominant tree species in southern China based on Sentinel-2 time series B. Yang et al. https://doi.org/10.1016/j.jenvman.2025.126293
- Switcher-HNet: A switchable hierarchical network for tree species classification from forest stand to individual tree tasks S. Liu et al. https://doi.org/10.1016/j.isprsjprs.2025.10.040
- Species information in multi-temporal Sentinel-2 data improves forest canopy height estimation C. Choi et al. https://doi.org/10.1016/j.agrformet.2026.111114
- Advances in tree detection and species classification using remote sensing data: a review V. Yilmaz https://doi.org/10.1080/14498596.2026.2658529
- Machine learning-based geospatial assessment of forest structure characteristics and sequestration potential for informed carbon stocks inventories U. Tasuev et al. https://doi.org/10.1038/s41598-026-50929-w
- Development of transferable hybrid deep learning networks for temporal and multi-regional mapping of poplar plantations with Sentinel-2 M. Ozturk & I. Colkesen https://doi.org/10.1016/j.asr.2025.07.075
- Classification of Dominant Tree Species and Vegetation Types Based on Sentinel-2 Time-Series Multispectral Data in the Eastern Qilian Mountains J. Yu et al. https://doi.org/10.1109/JSTARS.2025.3646463
15 citations as recorded by crossref.
- Implementing a growth pattern of climax community as post-processing filters core to improve tree species classification accuracy B. Zhang et al. https://doi.org/10.1080/17538947.2025.2498601
- Multi-Temporal Fusion of Sentinel-1 and Sentinel-2 Data for High-Accuracy Tree Species Identification in Subtropical Regions H. Li et al. https://doi.org/10.3390/rs18040592
- Integrating ChloroNet and XAI for accurate SPAD prediction in rice M. Abbasi et al. https://doi.org/10.1016/j.rsase.2026.101971
- Combining Sentinel-2 time-series and SITS-BERT for area-wide mapping of eight central European tree species J. Költzow et al. https://doi.org/10.1093/forestry/cpag042
- Geometry-Constrained Reference Sample Construction from Forest Inventory Compartments for Dominant Tree Species Mapping P. Zheng et al. https://doi.org/10.3390/rs18172915
- Multi-temporal Sentinel-2 images and LiDAR data fusion in dominant tree species classification D. Rim & I. Farah https://doi.org/10.1007/s41060-026-01218-2
- Identifying (less) vulnerable forest structures to storms in different conifer-dominated European forest landscapes by simple spaceborne proxies B. Garamszegi et al. https://doi.org/10.1139/cjfr-2025-0115
- Predicting rice crop height from field and Sentinel-2 data M. Abbasi et al. https://doi.org/10.1080/01431161.2025.2582213
- Large-scale mapping and uncertainty assessment of dominant tree species in southern China based on Sentinel-2 time series B. Yang et al. https://doi.org/10.1016/j.jenvman.2025.126293
- Switcher-HNet: A switchable hierarchical network for tree species classification from forest stand to individual tree tasks S. Liu et al. https://doi.org/10.1016/j.isprsjprs.2025.10.040
- Species information in multi-temporal Sentinel-2 data improves forest canopy height estimation C. Choi et al. https://doi.org/10.1016/j.agrformet.2026.111114
- Advances in tree detection and species classification using remote sensing data: a review V. Yilmaz https://doi.org/10.1080/14498596.2026.2658529
- Machine learning-based geospatial assessment of forest structure characteristics and sequestration potential for informed carbon stocks inventories U. Tasuev et al. https://doi.org/10.1038/s41598-026-50929-w
- Development of transferable hybrid deep learning networks for temporal and multi-regional mapping of poplar plantations with Sentinel-2 M. Ozturk & I. Colkesen https://doi.org/10.1016/j.asr.2025.07.075
- Classification of Dominant Tree Species and Vegetation Types Based on Sentinel-2 Time-Series Multispectral Data in the Eastern Qilian Mountains J. Yu et al. https://doi.org/10.1109/JSTARS.2025.3646463
Saved (final revised paper)
Latest update: 07 Sep 2026
Short summary
Classifying tree species in satellite images is an important task for environmental monitoring and forest management. Here we present a dataset containing Sentinel-2 satellite pixel time series of individual trees intended for training machine learning models. The dataset was created by merging information from the German National Forest Inventory in 2012 with satellite data. It sparsely covers the whole of Germany for the years 2015 to 2022 and comprises 48 species and 3 species groups.
Classifying tree species in satellite images is an important task for environmental monitoring...
Altmetrics
Final-revised paper
Preprint