Above-Ground Phenotyping of Sugi Using Terrestrial Lidar

For tree improvement program of Cryptomeria japonica (sugi in Japanese), a number of important above-ground traits (tree height, DBH, stem bending, stem curve, crown depth, etc.) were measured using point cloud data obtained by terrestrial LiDAR (Light Detection and Ranging). For three-dimension scanning, a F₁ progeny test site of C. japonica established in Forest Tree Breeding Center, FFPRI (Ibaraki, Japan) was used. First, 2m-mesh DEM (Digital Elevation Model) was constructed from point cloud data. Second, each individual tree stem was detected from point clouds automatically by a clue for the presence of points per set height. Third, horizontally aligned points which were assumed as the stem surface per 10cm height from 30cm above-ground were approximated to true circles. As a result, almost all individuals within the surveyed area were detected successfully. Coefficient of determination between measured values and estimated values of tree height and DBH were very high (R²= 0.903 and 0.975, respectively). Based on the data of approximated circle, stem bending and stem curve were evaluated. Crown depth was also estimated accurately from the circle data. Furthermore, genetic parameters for these traits were calculated. Using point cloud data, it is possible to obtain a large amount of information of traits of forest trees at the same time. This phenotyping procedure will be useful not only for conventional tree breeding but also for genome wide studies.