Evaluating airborne, mobile and terrestrial laser scanning for urban tree inventories: A case study in Ghent, Belgium

Abstract

In urban tree inventories, structural measurements such as diameter at breast height (DBH), tree height (H), crown projection area (CPA) and crown volume (CV) are essential for diverse applications, including accurate ecosystem service estimation and management decisions. Traditionally, tree measurements are obtained using range finders and diameter tape. These measurements can be integrated into urban tree inventories through 3D laser scanning (also known as LiDAR). Multiple platforms for laser scanning exist, each with its own advantages and disadvantages, which have not yet been compared explicitly for urban tree inventories. We collected terrestrial laser scanning (TLS), mobile laser scanning (MLS) and airborne laser scanning (ALS) in leaf-on (TLS, MLS, ALS) and leaf-off conditions (TLS, MLS) in Ghent, Belgium. We evaluated the DBH, H, CPA and CV of 95 individual trees acquired from each acquisition platform, benchmarking against TLS. Our results show accurate DBH derivation from both TLS and MLS (bias < 2 cm, concordance correlation coefficient (CCC) ≈ 1). However, during leaf-on conditions, occlusion from shrubs and ivy is observed. For leaf-on MLS, point clouds of large trees exhibited occlusion in the top canopy, impacting crown volume (CV MLS leaf-on: bias = −116 m², CCC=0.85) and, to a lesser extent, tree height (H MLS leaf-on: bias = −0.38 m, CCC=0.99). Crown projected area was less affected (bias = 0.49 m², CCC=0.99), with differences more attributed to varying point precision among sensors. The difference between the metric and benchmark increased with tree size and the structural complexity of the surroundings (e.g. buildings), especially for MLS, for which limited GNSS coverage, traffic, and suboptimal walking patterns impeded ideal data collection. Our results will help city councils and tree managers choose the most optimal LiDAR platform for urban tree inventories, accounting for their purpose, site complexity and budget.