Understanding the forest structure: development of tools for identification and delineation of individual trees using LiDAR.

• Main Author: Loos, Rafael
• Other Authors: Niemann, K. O.
• Language: English; en
• Published: 2010
• Subjects: Trees; Forest canopies; Forest ecology; UVic Subject Index::Sciences and Engineering::Agriculture::Forests and forestry
• Online Access: http://hdl.handle.net/1828/3110

LiDAR (Light Detection and Ranging) is currently being used to extract the biophysical characteristics of forests. LiDAR can provide extensive information about tree canopies; pulses reflected back to the sensor can represent understory vegetation as well as partial tree canopies below the dominant trees. Canopy structure can yield valuable clues about the biodiversity, and processes affecting the ecology of forest stands. In addition, structural information can provide insight into other processes such as fire behaviour and the distribution of fuels. This thesis focuses in developing tools to better understand the forest structure. The tools are computer-based algorithms that use LiDAR data as input and provide output with detailed information about the different layers of vegetation in a forested area. Three main modules are used in this study: (1) treetop identification, (2) delineation of canopies for the dominant layer of vegetation, and (3) delineation of partial canopies underneath this dominant layer. The study area was located in the Greater Victoria Water District, west of Victoria, British Columbia, Canada. Nine plots were chosen to represent the study area. A complete census was conducted in the summer of 2005 to provide information about tree, diameter at breast height (DBH), and tree dominance (based on the criteria: suppressed, intermediate, codominant and dominant). Results show that the algorithm is able to properly identify and delineate the majority of trees in the study areas. The third module, partial canopy delineation, also presents promising results with the dataset used. Newer LiDAR systems, with higher number of returns, will definitely provide better datasets with more information of the different layers within the forest, increasing the identification and delineation of these partial trees. Need of new field data is a must for future work and for further tests with the algorithm.