Spatial Transferability of Machine Learning Models for Urban Tree Species Classification using Sentinel-2 Multispectral Imagery

Abstract

Urban tree species classification supports urban forest management, biodiversity mapping, and assessments of the urban environment. Machine learning models have shown promise for urban species classification using remote sensing data, but their spatial transferability remains unexplored in urban environments. This study evaluates the spatial transferability of Extreme Gradient Boosting (XGBoost), a classical machine learning algorithm, for urban species classification using Sentinel-2 imagery. Fifteen tree species common to New York, New York; Portland, Oregon; and Seattle, Washington were selected from tree inventories. Time-series Sentinel-2 Level-2A surface reflectance imagery and NDVI data from 2024 were used to classify samples. Random oversampling and a unified sample size were both tested. Using random oversampling, on-site models were moderately successful, achieving overall accuracies of 0.50 to 0.58 and kappa values of 0.40 to 0.51, although accuracy may be attributed to class imbalance. When models were transferred to use testing data from alternate study areas, overall accuracy and kappa values decreased significantly, to 0.10-0.24 and 0.04-0.12, respectively. Using balanced data, models achieved overall accuracies of 0.21 to 0.43, and kappa values of 0.16 to 0.39. Overall accuracy decreased to 0.07 to 0.16, and kappa values decreased to 0.01 to 0.10. The results show the limited spatial transferability of classical machine learning models for urban tree species classification when using Sentinel-2 data, due to severe class imbalances, amplified through model transfer, phenological and data availability differences, and spatial context. This study emphasizes the benefits of localized training data for urban tree species classification and highlights the decrease in accuracy for transferred urban tree species classification models using Sentinel-2 imagery.

Keywords

• Remote sensing
• Tree species classification
• Machine learning
• Sentinel-2
• XGBoost

Ethical Statement

The authors declare that there is no conflict of interest.

References

1. Zhong, L., Dai, Z., Fang, P., Cao, Y., & Wang, L. (2024). A Review: Tree Species Classification Based on Remote Sensing Data and Classic Deep Learning-Based Methods. Forests, 15(5), 852. https://doi.org/10.3390/f15050852
2. Fassnacht, F. E., Latifi, H., Stereńczak, K., Modzelewska, A., Lefsky, M., Waser, L. T., … Ghosh, A. (2016). Review of studies on tree species classification from remotely sensed data. Remote Sensing of Environment, 186, 64–87. https://doi.org/10.1016/j.rse.2016.08.013
3. Pu, R. (2021). Mapping Tree Species Using Advanced Remote Sensing Technologies: A State-of-the-Art Review and Perspective. Journal of Remote Sensing, 2021. https://doi.org/10.34133/2021/9812624
4. Ocón, J. P., Stavros, E. N., Steinberg, S. J., Robertson, J., & Gillespie, T. W. (2024). Remote sensing approaches to identify trees to species-level in the urban forest: A review. Progress in Physical Geography: Earth and Environment, 48(3), 438–453. https://doi.org/10.1177/03091333241252520
5. Persson, M., Lindberg, E., & Reese, H. (2018). Tree Species Classification with Multi-Temporal Sentinel-2 Data. Remote Sensing, 10(11), 1794. https://doi.org/10.3390/rs10111794
6. Li, D., Ke, Y., Gong, H., & Li, X. (2015). Object-Based Urban Tree Species Classification Using Bi-Temporal WorldView-2 and WorldView-3 Images. Remote Sensing, 7(12), 16917-16937. https://doi.org/10.3390/rs71215861
7. Liu, P., Ren, C., Wang, Z., Jia, M., Yu, W., Ren, H., & Xia, C. (2024). Evaluating the Potential of Sentinel-2 Time Series Imagery and Machine Learning for Tree Species Classification in a Mountainous Forest. Remote Sensing, 16(2), 293. https://doi.org/10.3390/rs16020293
8. Lim, J., Kim, K.-M., & Jin, R. (2019). Tree Species Classification Using Hyperion and Sentinel-2 Data with Machine Learning in South Korea and China. ISPRS International Journal of Geo-Information, 8(3), 150. https://doi.org/10.3390/ijgi8030150

9. Sun, Y., Huang, J., Ao, Z., Lao, D., & Xin, Q. (2019). Deep Learning Approaches for the Mapping of Tree Species Diversity in a Tropical Wetland Using Airborne LiDAR and High-Spatial-Resolution Remote Sensing Images. Forests, 10(11), 1047. https://doi.org/10.3390/f10111047
10. Kirby, K. R., & Potvin, C. (2007). Variation in carbon storage among tree species: Implications for the management of a small-scale carbon sink project. Forest Ecology and Management, 246(2–3), 208–221. https://doi.org/10.1016/j.foreco.2007.03.072
11. Arasumani, M., Bunyan, M., & Robin, V. V. (2021). Opportunities and challenges in using remote sensing for invasive tree species management, and in the identification of restoration sites in tropical montane grasslands. Journal of Environmental Management, 280, 111759. https://doi.org/10.1016/j.jenvman.2020.111759
12. Boschetti, M., Boschetti, L., Oliveri, S., Casati, L., & Canova, I. (2007). Tree species mapping with Airborne hyper‐spectral MIVIS data: the Ticino Park study case. International Journal of Remote Sensing, 28(6), 1251–1261. https://doi.org/10.1080/01431160600928542
13. Abdollahnejad, A., & Panagiotidis, D. (2020). Tree Species Classification and Health Status Assessment for a Mixed Broadleaf-Conifer Forest with UAS Multispectral Imaging. Remote Sensing, 12(22), 3722. https://doi.org/10.3390/rs12223722
14. Baldeck, C. A., Colgan, M. S., Féret, J.-B., Levick, S. R., Martin, R. E., & Asner, G. P. (2014). Landscape‐scale variation in plant community composition of an African savanna from airborne species mapping. Ecological Applications, 24(1), 84–93. https://doi.org/10.1890/13-0307.1
15. Ossola, A., & Hopton, M. E. (2018). Measuring urban tree loss dynamics across residential landscapes. Science of The Total Environment, 612, 940–949. https://doi.org/10.1016/j.scitotenv.2017.08.103
16. Jensen, R. R., Hardin, P. J., & Hardin, A. J. (2012). Classification of urban tree species using hyperspectral imagery. Geocarto International, 27(5), 443–458. https://doi.org/10.1080/10106049.2011.638989
17. Laverne, R. J., & Winson-Geideman, K. (2003). The Influence of Trees and Landscaping On Rental Rates at Office Buildings. Arboriculture & Urban Forestry, 29(5), 281–290. https://doi.org/10.48044/jauf.2003.032
18. Jensen, R., Gatrell, J., Boulton, J., & Harper, B. (2004). Using Remote Sensing and Geographic Information Systems to Study Urban Quality of Life and Urban Forest Amenities. Ecology and Society, 9(5). https://doi.org/10.5751/es-01201-090505
19. Dwyer, J., McPherson, E. G., Schroeder, H., & Rowntree, R. (1992). Assessing the Benefits and Costs of the Urban Forest. Arboriculture & Urban Forestry, 18(5), 227–234. https://doi.org/10.48044/jauf.1992.045
20. Graves, S. J., Marconi, S., Stewart, D., Harmon, I., Weinstein, B., Kanazawa, Y., … White, E. P. (2023). Data science competition for cross-site individual tree species identification from airborne remote sensing data. PeerJ, 11, e16578. https://doi.org/10.7717/peerj.16578
21. Gascon, F., Bouzinac, C., Thépaut, O., Jung, M., Francesconi, B., Louis, J., Lonjou, V., Lafrance, B., Massera, S., Gaudel-Vacaresse, A., Languille, F., Alhammoud, B., Viallefont, F., Pflug, B., Bieniarz, J., Clerc, S., Pessiot, L., Trémas, T., Cadau, E., ... Fernandez, V. (2017). Copernicus Sentinel-2A Calibration and Products Validation Status. Remote Sensing, 9(6), 584. https://doi.org/10.3390/rs9060584
22. Gatti, A., & Bertolini, A. (2015). Sentinel-2 products specification document. Rapport technique, 4-7.
23. Huang, S., Tang, L., Hupy, J. P., Wang, Y., & Shao, G. (2020). A commentary review on the use of normalized difference vegetation index (NDVI) in the era of popular remote sensing. Journal of Forestry Research, 32(1), 1–6. https://doi.org/10.1007/s11676-020-01155-1
24. Huete, A., Didan, K., Miura, T., Rodriguez, E. P., Gao, X., & Ferreira, L. G. (2002). Overview of the radiometric and biophysical performance of the MODIS vegetation indices. Remote Sensing of Environment, 83(1–2), 195–213. https://doi.org/10.1016/s0034-4257(02)00096-2
25. City of Seattle. (2025, August 2). SDOT Trees [Data set]. Retrieved August 7, 2025, from https://catalog.data.gov/dataset/sdot-trees
26. Portland Parks & Recreation (Bureau of Parks & Recreation, City of Portland). (n.d.). *Street Tree Inventory – Active Records* [Data set]. Retrieved August 7, 2025, from https://gis‑pdx.opendata.arcgis.com/datasets/street‑tree‑inventory‑active‑records
27. City of New York, Department of Parks and Recreation. (2024, November 15). 2015 Street Tree Census – Tree Data [Data set]. Retrieved August 7, 2025, from https://catalog.data.gov/dataset/2015-street-tree-census-tree-data-a16a1
28. Chen, T., & Guestrin, C. (2016). XGBoost. ACM. https://doi.org/10.1145/2939672.2939785
29. Congalton, R. G., & Green, K. (2019). Assessing the Accuracy of Remotely Sensed Data. CRC Press. https://doi.org/10.1201/9780429052729
30. Orynbaikyzy, A., Gessner, U., & Conrad, C. (2022). Spatial Transferability of Random Forest Models for Crop Type Classification Using Sentinel-1 and Sentinel-2. Remote Sensing, 14(6), 1493. https://doi.org/10.3390/rs14061493
31. Liu, P., Ren, C., Wang, Z., Jia, M., Yu, W., Ren, H., & Xia, C. (2024). Evaluating the Potential of Sentinel-2 Time Series Imagery and Machine Learning for Tree Species Classification in a Mountainous Forest. Remote Sensing, 16(2), 293. https://doi.org/10.3390/rs16020293
32. Mouret, F., Albughdadi, M., Duthoit, S., Kouamé, D., Rieu, G., & Tourneret, J.-Y. (2022). Reconstruction of Sentinel-2 derived time series using robust Gaussian mixture models — Application to the detection of anomalous crop development. Computers and Electronics in Agriculture, 198, 106983. https://doi.org/10.1016/j.compag.2022.106983
33. DEMİR, S., & ŞAHİN, E. K. (2022). Evaluation of Oversampling Methods (OVER, SMOTE, and ROSE) in Classifying Soil Liquefaction Dataset based on SVM, RF, and Naïve Bayes. European Journal of Science and Technology. https://doi.org/10.31590/ejosat.1077867