High terrestrial laser scanning: An accessible method for tree crown characterization

We introduce high Terrestrial Laser Scanning (hTLS), a method that enhances the assessment of single-tree metrics and Quantitative Structure Models (QSM) on forest monitoring plots by using an 8-m-tall telescopic tripod. Terrestrial Laser Scanning (TLS) has proven useful in research and forest monitoring but has limitations in fully capturing the complexity of upper tree crowns of tall trees. This pilot study analyses the impact of an elevated scanning position on single-tree metrics (tree height, crown length, crown projection area, convex hulls of the tree), QSMs and vertical voxel count distributions of 40 leaf-off European beech trees ( Fagus sylvatica ) in Germany. We used the same single-tree metrics derived from the QSMs as proxy for reconstruction success. Our results show that hTLS increases point density by 43.32%, especially in the upper crown regions of trees taller than 25 m. While the impact on point cloud-based single-tree metrics is found significant, it is often small. We found Concordance Correlation Coefficients (CCC) between our reference and hTLS ranging from 0.962 to 1 and for TLS ranging from 0.926 to 0.999. The impact of hTLS is more pronounced for the derivation of QSMs. The achieved CCCs ranged between 0.593 and 0.998 for metrics derived from hTLS-based QSMs and between 0.457 and 0.996 for metrics derived from TLS-based QSMs. The elevated scanning position mitigated occlusion, decreased measurement error and allowed to reconstruct more branches, especially in the upper tree crown of trees above 25 m height. • Our method provides better scanning positions for crown metrics and reconstructions. • Using a tall carbon-fiber tripod (8 m) increases point density at the treetops. • The positive impact vs. conventional near-ground scans increases with taller trees. • Especially Quantitative Structure Models benefit from an elevated scanning position.