Using data from Laser Scanning and Cyclic Aerial Photography of Slovenia, we i) assessed the structure of forest stands and demonstrated different methods for segmentation of individual trees, ii) assessed the possibilities of detecting changes in the stand structure and iii) assessed the influence of physiographic factors on differences in stand structure. Using three algorithms, we performed the segmentation of individual trees on five sample plots. As expected, the laser scan data gave better results than the cyclic aerial survey data. The best results were achieved in an artificially established stand with a homogeneous stand structure, where we correctly identified 75% of the trees. In adult stands with uniform structure, we achieved 26-32% of correctly identified trees, while in stands with uneven-aged structure, we correctly identified 30-41% of all trees. We created a map of changes for the period 2015 – 2020 for the selected forest sections and estimated the changes in stand structure to be 9.7% of the area in relation to the entire area of the selected forest sections. The influence of physiographic factors on the stand structure was explained with the help of a topographic position index. The results showed that trees with a height of more than 40 m appear in the sections where distinct ditches appear. There are also differences in average tree heights depending on the topographical position.
|