Optimizing line-plot size for personal laser scanning: modeling distance-dependent tree detection probability along transects

Abstract

Personal laser scanning (PLS) systems are gaining popularity in forest inventory research and practice. They are primarily utilized on circular or compact rectangular sample plots to mitigate potential instrument drift and enhance tree detection rates, and a closed-loop scan path is usually implemented to achieve these objectives, ensuring thorough coverage of the plot. This study introduced a novel approach by applying the distance-sampling framework to PLS data collected during walks along line transects. Modeling the distance-dependent probability of tree detection using PLS coupled with automatic routines for point cloud processing aimed to ascertain the optimal width of line-plots to maximize tree detection rates. The optimized plots exhibited tree detection rates exceeding 99%, which facilitated accurate estimates of tree density, basal area, and growing stock volumes. This proposed method demonstrated considerable potential for data collection while walking along line transects in forests. For instance, the otherwise unproductive working time of field crews moving between systematically arranged sample plots can be utilized for additional data collection without generating additional costs. This innovative approach not only enhances operational efficiency but also establishes a foundation for further advancements to explore PLS applications in forest management practices.

Keywords

• Personal Laser Scanning
• Lidar
• Forest Inventory
• Distance Sampling
• Line Transect Sampling
• Tree Detection