Detection of European Aspen (Populus tremula L.) Based on an Unmanned Aerial Vehicle Approach in Boreal Forests

European aspen (Populus tremula L.) is a keystone species for biodiversity of boreal forests. Large-diameter aspens maintain the diversity hundreds species, many which are threatened in Fennoscandia. Due to low economic value and relatively sparse scattered occurrence forests, there lack information spatial temporal distribution aspen, hampers efficient planning implementation sustainable forest management practices conservation efforts. Our objective was assess identification at individual tree level southern using high-resolution photogrammetric point cloud (PPC) multispectral (MSP) orthomosaics acquired with an unmanned aerial vehicle (UAV). The structure-from-motion approach applied generate RGB imagery-based PPC be used tree-crown delineation. Multispectral data were collected two UAV cameras: Parrot Sequoia MicaSense RedEdge-M. Tree-crown outlines obtained from watershed segmentation intersected mosaics extract calculate spectral metrics trees. We assessed role features extracted combination it, machine learning classifier—Support Vector Machine (SVM) perform different classifications: discrimination other combined into one class classification all four (aspen, birch, pine, spruce) simultaneously. In first scenario, highest accuracy 84% (F1-score) overall 90.1% achieved only PPC, whereas second 86 % 83.3% MSP features. proposed method provides new possibility rapid assessment enable more as well contribute monitoring efforts