Challenges to Estimating Tree Height via LiDAR in Closed-Canopy Forests: A Parable from Western Oregon

摘要

Accuracy of LiDAR-derived tree height estimates obtained from high return density LiDAR data acquired at leaf-on and leaf-off conditions was evaluated in the steep-sloped, high canopy cover forests of western Oregon using precise field measurements obtained with survey-grade equipment. Overall, LiDAR error exceeded 1 0% of tree height for 60% of the trees and for 43% of the plots at leaf-on and for 55% of the trees and for 38% of the plots at leaf-off. These errors, which are much greater than what has been typically reported, can be decomposed into three sources: tree-top location, ground elevation model, and tree lean. To obtain accurate LiDAR-derived estimates of tree height in conditions similar to those in western Oregon would require revision of LiDAR acquisition specifications and further investment in analysis techniques. [PUBLICATION ABSTRACT]