Close-range laser scanning in forests: towards physically based semantics across scales

F. Morsdorf, D. Kükenbrink, F. D. Schneider, M. Abegg, M. E. Schaepman
2018 Interface Focus  
One contribution of 12 to a theme issue 'The terrestrial laser scanning revolution in forest ecology'. Laser scanning with its unique measurement concept holds the potential to revolutionize the way we assess and quantify three-dimensional vegetation structure. Modern laser systems used at close range, be it on terrestrial, mobile or unmanned aerial platforms, provide dense and accurate threedimensional data whose information just waits to be harvested. However, the transformation of such data
more » ... o information is not as straightforward as for airborne and space-borne approaches, where typically empirical models are built using ground truth of target variables. Simpler variables, such as diameter at breast height, can be readily derived and validated. More complex variables, e.g. leaf area index, need a thorough understanding and consideration of the physical particularities of the measurement process and semantic labelling of the point cloud. Quantified structural models provide a framework for such labelling by deriving stem and branch architecture, a basis for many of the more complex structural variables. The physical information of the laser scanning process is still underused and we show how it could play a vital role in conjunction with three-dimensional radiative transfer models to shape the information retrieval methods of the future. Using such a combined forward and physically based approach will make methods robust and transferable. In addition, it avoids replacing observer bias from field inventories with instrument bias from different laser instruments. Still, an intensive dialogue with the users of the derived information is mandatory to potentially re-design structural concepts and variables so that they profit most of the rich data that closerange laser scanning provides.
doi:10.1098/rsfs.2017.0046 pmid:29503725 fatcat:durvze7zcze33jotfspy4xx4ka