DroneNet4Beetles - A shared drone data initiative for the early detection of European spruce bark beetle outbreaks - Level 2 data: Crown Reflectance

The datasets published here were assembled within the DroneNet4Beetles initiative (https://www.slu.se/dronenet4beetles), an open, collaborative effort to harmonize multi-site UAV data for the early detection of spruce bark beetle infestation. The data version V1.0.0 particularly contains the data used in the study “A cross-European assessment on the pre-emergence detection of trees attacked by spruce bark beetle using UAV imagery” in Remote Sensing of Environment, https://doi.org/10.1016/j.rse.2026.115445 The dataset supports a multi-country benchmark for detecting Norway spruce (Picea abies) trees attacked by the European spruce bark beetle (Ips typographus). Data were assembled from Sweden, Finland, Italy, and the Czech Republic, representing Nordic and Central European outbreak conditions. The archive includes six UAV remote-sensing datasets: SE2021-MS, SE2023-MS, SE2023-HS, FI2023-MS, IT2023-MS, and CZ2022-HS. Together, these datasets include crown-averaged spectral reflectance from multispectral and hyperspectral UAV images and tree-level health-status information used in the paper. The remote-sensing data include 26 multispectral and 16 hyperspectral image sets acquired using MAIA S2, SPECIM AFX10, MicaSense Altum-PT, DJI Phantom 4 Multispectral, and Headwall Nano-Hyperspec sensors. The datasets cover multiple acquisition dates during the growing season and include both pheromone-induced and naturally occurring bark beetle attacks. Tree-level reference data distinguish healthy and infested trees based on field inventory or image interpretation, depending on the study site. The full dataset covers 1,798 infested and 13,020 healthy trees, including 12 study areas and six time series. The archive is intended to support reproducibility of the published analysis and to provide benchmark data for future studies on early detection of bark beetle infestation, UAV-based forest health monitoring, vegetation-index transferability, sensor comparison, and single-tree spectral analysis. The data can be used to evaluate how detection performance depends on sensor type, band availability, red-edge wavelength, green-shoulder sensitivity, vegetation-index normalization, local decline dynamics, outbreak phase, and site conditions. Dataset contents Each dataset folder contains tree-level ground-truth information, including stand number, plot number, and tree status classified as Healthy or Infested. It also contains crown-level reflectance data, where each row corresponds to an individual tree and follows the same order as the ground-truth file, while each column represents reflectance at Band1 to Bandn, as defined by the sensor. When a tree crown was not successfully segmented from the images, or when images from a given band were unavailable, the corresponding reflectance value was recorded as NaN. For the hyperspectral datasets SE2023-HS and CZ2022-HS, an additional wavelength file is provided to indicate the wavelength corresponding to each spectral band. More details of the data are presented in the paper (https://doi.org/10.1016/j.rse.2026.115445) Reuse statement These data are suitable for research and benchmarking related to UAV remote sensing of forest health, early detection of bark beetle infestation, vegetation-index development, hyperspectral and multispectral sensor comparison, crown-level spectral analysis, and reproducibility of the associated publication. Users should cite both this Zenodo dataset and the associated Remote Sensing of Environment paper (https://doi.org/10.1016/j.rse.2026.115445) when using the data. Huo, L., Cosimo, L. H. E., Bozzini, A., Bijou, S., Alves de Oliveira, R., Suomalainen, J., Vennervirta, E., Koivumäki, N., Näsi, R., Kupková, L., Faccoli, M., & Honkavaara, E. (2026). A cross-European assessment on the pre-emergence detection of trees attacked by spruce bark beetle using UAV imagery. Remote Sensing of Environment, 115445. https://doi.org/10.1016/j.rse.2026.115445 Join the future data network https://www.slu.se/dronenet4beetles