Comparison of Methodological Approaches to Measuring Trees' Heights and Crown Projections Based on the Laser Scanning Data from an Unmanned Aerial Vehicle in the Middle Taiga Forests

Boreal forests are the most important reservoirs of terrestrial carbon. Their wood makes up to 30-35% of their overall carbon stock, making their inventory a task of national importance. The world practice of recent decades is scaling the forest inventory data using unmanned aerial vehicles (UAVs), which are an effective and inexpensive means of collecting information at the level of individual biogeocoenoses. The variety of survey and processing methods requires finding optimal approaches to wood stock inventory using UAVs. The article is dedicated to comparing two methods (Matlab - ML and TerraScan - TS) of estimating the number, height and projective cover of tree crowns by processing point clouds of laser reflections of different densities (100 and 2800 points / m; ML, ML and TS, TS). Lidar survey was carried out on three 50 × 50 metres sample plots. SP 1 - forest crossed by a stream; SP 2 - forest on a flat surface; SP 3 - forest bordering an oligotrophic swamp. The number of trees identified using ML / ML for SPs 1, 2 and 3 was: 55 / 66, 67 / 87 and 174 / 220; while using TS and TS resulted in following numbers: 66 / 95, 85 / 156 and 82 /166 respectively. The average tree height reaches 10-24 m, with little difference between the two processing methods at each site. The total projective crown cover for SPs 1, 2 and 3 was 2569, 2494 and 2059 m in ML; 2091, 2424 and 1506 m in TS. During remote assessment, only the first storey trees’ number could be estimated with a sufficient degree of completeness: for this task, the ML method was sufficient. However, for a more comprehensive inventory of trees, including lower storeys, it is preferable to use TS, under the condition of an adequate selection of the minimum tree height parameter’s value.