Volumetric overestimation of small branches in 3D reconstructions of <i>Fraxinus excelsior</i>

Terrestrial laser scanning (TLS) has been applied to estimate forest wood volume based on detailed 3D tree reconstructions from point cloud data. However, sources of uncertainties in the data (alignment and scattering errors, occlusion, foliage...) reconstruction algorithm type parameterisation are known affect reconstruction, especially around finer branches. To better understand impacts these accuracy TLS-derived woody volume, high-quality TLS scans were collected leaf-off conditions prior destructive harvesting two forest-grown common ash trees ( L.; diameter at breast height ~28 cm, 732 868 L). We manually measured branch diameters 265 locations trees. Estimates Quantitative Structure Models (QSM) compared with manual measurements. The QSM estimates decreased smaller diameters. Tree was overestimated (+336 L +392 L) both Branches measuring < 5 cm accounted for 80% 83% this overestimation respectively. Filtering errors or improved coregistration approximately halved overestimation. Range filtering modified layouts had mixed effects. small overestimations originated primarily limitations scanner characteristics rather than suboptimal parameterisation. For a higher quality allows branches be accurately reconstructed. Additional experiments need elucidate if results can generalised beyond setup study.Fraxinus excelsior