Application of terrestrial LiDAR for forest tree breeding: Evaluation of stem traits on F₁ Progenies of Cryptomeria japonica plus trees

In order to apply terrestrial LiDAR for phenotyping in forest tree breeding, various traits of stem were quantified and evaluated based on a point cloud obtained by ground-based three-dimensional laser measurements on a study site of F₁ individuals of Cryptomeria japonica plus trees. The point cloud of each individual stem was approximated by "stem circles", which are horizontal perfect circles arrayed per 10 cm height. Based on the point cloud and the stem circles, both height and diameter at breast height (DBH) of individuals were automatically estimated with high accuracy. The relative stem taper curve equation of the third order was approximated for each individual based on the stem circles. Stem volume of each individual estimated from the taper curve showed high correlation with one estimated from only tree height and DBH. Stem straightness was quantified in the two standardized methods and these methods were evaluated with stem straightness index investigated on the field. The relationships among these estimated traits were examiend using correlation coefficients. Of the ratio of additive genetic variance, the highest was stem straightness, followed in order by tree height, stem volume, DBH, and the coefficients of the relative stem taper curve. The usefulness of the analytical method for phenotyping in forest tree breeding was discussed.