Estimating Mean Height and Stand Volume in Broad Leaved Forest Stands using LiDAR(<Special Issue>Silvilaser)

概要

論文の詳細を見る
Forest stand variables (mean height, stand volume, and mean diameter breast height (DBH), tree density) were estimated in evergreen and deciduous broad leaved forest stand using LiDAR. LiDAR data were acquired along 12km and 28km transects with 1 pulse per square meter and small foot print (20cm). We set plots in evergreen and deciduous broad leaved forest stands from small to large on the transect and measured forest stand variables (mean DBH: 3.4-41.2cm; mean H: 3.1-17.4m; V: 25.1-854m^3; N: 295-9,507ha^<-1>; n=18). Laser pulses of digital canopy height model were extracted in each plot and LiDAR indexes were calculated: average, maximum, 90, 75, 50, 25 percentiles, standard deviation, and coefficient of variation. A linear regression analysis was performed between LiDAR indexes and forest stand variables. Mean height had the highest relationship with the LiDAR index 75 percentile (r^2=0.79); stand volume with the LiDAR index average (r^2=0.79), mean DBH with the LiDAR index 75 percentile (r^2=0.56), and tree density with LiDAR index 75 percentile (r^2=0.52). These results showed that low density LiDAR was useful for forest stand variable and would be useful for update and modification of forest base map and forest register.