OPTIMIZING URBAN TREE ABOVE GROUND BIOMASS USING REMOTE SENSING AND TREE STRUCTURAL DIVERSITY

Above Ground Biomass (AGB) is vital to quantify the amount of carbon stored in terrestrial ecosystems and provides important information about local greenspaces and environments. Light Detection And Ranging (LiDAR) technology creates three-dimensional (3D) models of individual trees with centimeter-scale resolution. Quantitative Structure Models (QSMs) can produce a volume reconstruction for individual tree point clouds and calculate attributes such as height, Diameter at Breast Height (DBH), and total above-ground volume, hence an estimation of AGB. QSM is non-destructive and more accurate than traditional forestry practices which use generic allometric AGB equations. A set of parameters controls QSM’s volume reconstruction, and the best set varies for different tree structures. We collected in-field data such as DBH, total mass, and LiDAR scans on Midwestern State University’s campus. We optimized the parameters by comparing model outputs with in-field DBH and mass. We investigated the impact of tree structure on optimized parameters for each tree group based on their size. The optimized parameters improve the total volume estimation. QSM offers a better prediction of DBH and trunk volume. However, the total biomass estimation, based on the volume, produces some inaccuracy, possibly driven by many factors, such as distorted reconstruction of small branches, inaccurate estimation of wood density, scattered point clouds due to windy conditions, etc. Using the optimum parameters for individual tree groups, we estimated QSM-based AGB for all the trees. The outcome of this work will have real-world impacts for local stakeholders by informing them of the potential of LiDAR and data-based recommendations for carbon sinks and urban forest management strategies.