Quantifying distribution of above ground biomass estimates in forested areas across LiDAR applications and multiple scales

Abstract

Accurate quantification of forest biomass is essential for understanding carbon dynamics, preserving biodiversity, and advancing Nature-based Solutions (NbS) in Canada (Seddon et al., 2020). NbS harness, protect, or restore natural ecosystems to address complex environmental challenges. This thesis leverages LiDAR-derived point cloud data to estimate above-ground biomass (AGB), acknowledging that accuracy varies among different LiDAR sensors due to limitations like point density and survey altitude. We evaluated the effectiveness of three remote sensing technologies—Remotely Piloted Aircraft (RPA) LiDAR, aerial LiDAR, and Global Ecosystem Dynamics Investigation (GEDI) satellite LiDAR in estimating AGB across two distinct forested properties, Ballyduff Trails and Christie Bentham Wetland Trails. These remote sensing estimates were compared with traditional in-situ measurements to assess their accuracy and reliability.

Using statistical models and data processing algorithms, we derived AGB estimates from the LiDAR data. Results indicated that both RPA and Aerial LiDAR provided extensive spatial coverage but tended to underestimate AGB compared to in-situ measurements, likely due to the omission of smaller and understory trees. Aerial LiDAR measured higher average tree heights, effectively capturing upper canopy structures, while RPA LiDAR's higher point density facilitated detailed ground-level analysis. Despite its lower spatial resolution, GEDI satellite LiDAR offered a balance between coverage and accuracy, with estimates aligning closely with in-situ values for larger-scale assessments.

This study makes a novel contribution by being among the first to compare these three LiDAR technologies across different forest types in Canada. The findings have significant implications for improving forest management practices, informing conservation strategies, and enhancing carbon accounting methods. By highlighting the strengths and limitations of each remote sensing method, we provide valuable insights for their application in sustainable forest management and climate change mitigation initiatives.