ISSN 2756-3251
Advanced Journal of Environmental Science and Technology | Vol. 17 (6), pp. 646-658, June, 2026
Research Article
Multi-Spectral Drone Imagery and Convolutional Neural Networks Detect Early Bark Beetle Infestations in Temperate Coniferous Forests
Martin Gruber¹*, Aino Virtanen², Samuel Boateng³
¹Department of Forest and Soil Sciences, University of Natural Resources and Life Sciences, Vienna, Austria
²Department of Forest Sciences, University of Helsinki, Helsinki, Finland
³Department of Silviculture and Forest Management, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana
Corresponding Author: Martin Gruber — [email protected]
Abstract
Climate-driven droughts have triggered unprecedented outbreaks of European spruce bark beetle (Ips typographus), turning temperate forests from carbon sinks into carbon sources. We trained a 3D convolutional neural network on five-band multispectral UAV imagery (blue, green, red, red-edge, and near-infrared) collected biweekly across 1,200 hectares of Norway spruce forest. By detecting subtle reductions in red-edge chlorophyll indices prior to visible needle yellowing, the model identified green-attack stage trees with 91.8% overall accuracy and a 0.89 F1-score. Early sanitation felling guided by these maps reduced secondary timber volume losses by 63% compared to conventional ground surveys.
Keywords: bark beetle detection; multispectral drone imagery; convolutional neural networks; forest health monitoring; climate change adaptation; remote sensing
Manuscript Timeline: Received 19 May, 2026; Revised 16 June, 2026; Accepted 27 June, 2026; Published 30 June, 2026.
Citation: Gruber, M., Virtanen, A., and Boateng, S. (2026). Multi-Spectral Drone Imagery and Convolutional Neural Networks Detect Early Bark Beetle Infestations in Temperate Coniferous Forests. Advanced Journal of Environmental Science and Technology, 17(6), pp. 646-658.
Subscribe to read the full article: https://www.internationalscholarsjournals.org/subscribe-to-read