مجله جنگل ایران (Sep 2014)
Potential of LIDAR data for estimation of individual tree height of Acer velutinum and Carpinus betulus
Abstract
In traditional methods, measurement of tree height in broad-leaved forests, particularly in uneven-aged, high density and multi-storey stands is difficult and time-consuming. In the growing season of 2011, airborne laser scanner system was used to acquire Light Detection and Ranging (LIDAR) data by a sampling density of 4-5 points/m2 in a broad-leaved forest, northern Iran (Shast-Kalate forest located in southwest of Gorgan). LIDAR data was used to estimate the height of individual trees. Hundred individual tree samples belong to two species, Acer velutinum and Carpinus betulus, were selected. The trees were located in dominant storey and/or without overlay with adjacent trees. Heights of trees were measured using Vertex 1V. Center coordinates of the sample trees were determined using DGPS system and Total Station with distance and angle. Sample trees were detected in laser point clouds using their center coordinates and crown diameter. The heights of sample trees were extracted directly from laser point clouds. Linear regression model between the heights of trees in the field measurement and laser point clouds was yielded coefficients of determination (R2) (0.96, 0.95), RMSe (1.05m (6%), 1.48m (6%), and relative error (4%, 3%) for A.velutinum and C. betulus, respectively. Mean differences between tree height of laser point cloud and field measurement were achieved 1.04m and -1.53m for A.velutinum and C.betulus, respectively. The results clarified that considering the acceptable of LIDAR height data, rapid prevail of data and its easy access for users, the employment of LIDAR data for estimating tree height in other places of northern forests of Iran is advised.
