Estimating North Texas Urban Tree Above-Ground Biomass Based on Terrestrial LiDAR and Optimized Quantitative Structure Models

Abstract

Several studies have revealed that terrestrial light detection and ranging (LiDAR) remote-sensing technology can be an alternative, more accurate approach to estimate tree above-ground biomass (AGB). A set of algorithms has been developed to create a volume reconstruction of tree point clouds, which can be converted to an AGB estimate. These calculate AGB non-destructively and more accurately compared to current forestry practices, which use generic allometric equations. LiDAR scans of trees within Midwestern State University’s campus were collected and analyzed with a tree segmentation/modeling algorithm. The validation method was based on comparing the estimated above-ground attributes to actual field measurements. Optimized PD1 and PD2Max increase as the tree size increases, whereas PD2Min remains relatively the same for different tree sizes. Quantitative structure modeling (QSM) produces accurate diameter at breast height (DBH) estimates, however it fails to calculate it precisely when there are low branches or dense leaves within the canopy. These occlusions commonly occur with certain tree species, such as Pinus echinata and Juniperus virginiana. Our result suggests good agreement of QSM-derived AGB estimates for larger trees but overestimated AGB for smaller trees. This is due to the limitations in LiDAR technology, struggling to accurately scan fine branches and twigs of small trees, leading to errors in point cloud data and subsequent overestimation of their volume in the QSM. While the study provides valuable insights, the small sample size due to the complexity of destructive tree harvesting in urban ecosystems might limit the generalizability of the results.