Application of LiDAR-based DTMs and DSMs to detect landforms created by the tree uprooting process

Tree uprooting plays a significant role in shaping the microtopography in forested areas. This process leads to the formation of 1) pit-mound topography (with adjacent pit and mound forms being a result of the tree uprooting) and 2) root plates (with undecomposed root systems and soil material attached to them). The increasing availability and accuracy of LiDAR point clouds enable producing high-resolution DTMs and DSMs. Such models can be applied to detect even small forms with a diameter below 3 meters.
In the present project DTMs and DSMs were used to detect the location of 1) treethrow pit-mound pairs and 2) root plates of uprooted trees. Analysis was performed for selected 100x100 m research plots situated within the Babia Góra National Park (BgNP; Western Carpathians, Southern Poland). We used an open-source point cloud from the Polish Institute of Geodesy and Cartography (density: 20 pts/m2). All steps of the analysis were automated with the use of R programming language. Closed contour lines can be used to detect both types of forms. For pit-mound pairs detection, we tested different DTM resolutions and contour line intervals to achieve the best accuracy of the proposed contour method (CM). In the case of root plate detection, the point cloud was reduced to the points of the last return of the laser beam to maximize the chances of catching the points that actually reflect the locations of root plates and trunks. A differential model (DM) was produced by applying double classification of ground reflections with the use of a cloth simulation function algorithm (CSF). A contour line of the height of 1 m calculated from the DM was used to extract the potential locations of root plates.
The study has been supported by the Polish National Science Centre (project no 2019/35/O/ST10/00032).