Multidirectional laser measurement of three-dimensional tree structure

Abstract

In spite of the insistent need for comparable data, structural analysis of forest canopies with trees as their main components often is subjective, and thus insufficiently standardised. Current methods seldom are capable of describing the three-dimensional structure of a forest canopy, and the tree in particular, in an efficient, repeatable, accurate, complete and comprehensive manner. Laser technology provides a rising tool which creates the possibility to generate a unique and comprehensive mathematical description of deciduous tree structure. The terrestrial laser system used in this study consisted of the commercially available SICK Laser-Measurement-System-200 (LMS200). The laser scans the surrounding vegetation obtaining three-dimensional structural datasets with high resolution. The objective of this research was to investigate the influence of the geometric laser measurement pattern on the accuracy of the quantitative mathematical description of tree structure. Laser measurement variability along angular viewing differences is caused by the physical laser pulse/object interactions and an intrinsic characteristic of the laser device: a transmitted laser pulse is reflected by the first object it encounters. Therefore, spatial information, related to the position of the vegetative elements located behind the target object, was not available (shadow effect). These background objects had to be measured from different angles to obtain comprehensive laser coverage. An artificial tree was constructed in an experimental setup, thereby creating the possibility to arrange the structural key elements (leaves and branches) according to predetermined patterns. The multidirectional accuracy and quality of the laser measurement setups subsequently were compared based on the development of a mathematical description of each case.