Impact of the sampling design on the quality of ground-based LiDAR datasets

This research was undertaken to study the influence of the sampling design and laser beam density of ground-based LiDAR measurements on the quality of laser datasets in terms of shadowing. The generation of virtual forest stands by means of stochastic L-systems as tree descriptors were opted for based on the study frame. The dynamic plant modeler and plant nursery natFX (Bionatics, CIRAD, Montpellier, France) was used to simulate full grown forest stands of two tree species (i.e. Fagus sylvatic and Platanus acerifolia). Next, hemispherical laser measurements with different laser beam densities were simulated according to three different sampling patterns (i.e. single, diamond, and corners) inside these virtual forest stands through the use of ray-tracing technology. An adjusted sampling design has proven its effectiveness since an average decrease of 27.27% in shadowing in comparison with a single measurement was obtained. This contrasts with an average decrease of 13.64% by increase of the laser beams density by a factor 25. Afterwards, contact frequency values, calculated from the virtual laser data sets, were utilized to successfully model the shadowed parts of the canopy demonstrating the potential of ground-based laser scans to capture the 3D leaf distribution inside a full grown forest stand.