Investigating the structural condition of trees using LiDAR metrics

Unlike other investigations that use discrete return (DR) light detection and ranging (LiDAR) data for the visualisation and investigation of physical structure, this research attempts to investigate data relationships found within the LiDAR point cloud in order to infer the condition of the subject of interest, which for the purposes of this investigation are tree canopies. During DR LiDAR data capture, a laser pulse is emitted from the scanner and information about the subject is captured at only at specific points known as laser returns (r). Subsequently, not all of the laser waveform is recorded, meaning that although a general impression of the subject is captured in a point cloud, there will be areas of the subject between each data point that remain unrepresented in the dataset. This paper outlines preliminary research into attempting to discover what range of LiDAR metrics, and resulting data relationships, are the most suitable to identify the significance of the structural condition of tree canopies from a tree health perspective. This research contributes to a wider investigation of automated tree health assessment and the early identification of structural failure in trees using remote sensing techniques