USING TERRESTRIAL LASER SCANNING AND MODEL OPTIMIZATION FOR CHARACTERIZING URBAN TREE STRUCTURAL PARAMETERS

Above-ground biomass (AGB) is an important metric used to quantify the mass of carbon storage in terrestrial ecosystems. Urban trees have long been valued for the potential to store significant AGB. However, urban environments pose various challenges for accurately recording AGB due to the plasticity of tree form, high species diversity, and heterogeneous and complex land cover. Light Detection and Ranging (LiDAR) is a state-of-the-art technology of remote sensing that offers an opportunity to assess AGB in urban trees effectively and inexpensively. Using LiDAR and geometric modeling algorithms, we can generate three-dimensional (3D) point clouds and digital tree segmentations. These 3D models can be used to estimate urban tree structural parameters such as height, diameter at breast height, volume, surface area, etc. First, we manually extract individual tree point clouds from plot-scale LiDAR data collected in an urban setting at Midwestern State University (MSU Texas). We then apply state-of-the-art digital tree segmentation and geometric modeling algorithms to estimate tree structural parameters and AGB of these urban tree species. Finally, we validate the available modeling tools by optimizing parameters and comparing the outputs from the model simulations with extensive field measurements (harvested total biomass and branch biomass) from the campus trees. The model with optimized parameters for North Texas tree species (Red Oak and Cedar Elm) would provide a more accurate estimate of the total AGB of these urban tree species. This can offer a non-destructive approach for estimating urban tree AGB which is essential for understanding the overall atmospheric carbon sink and also has practical recommendations for sustainable urban forest management strategies.