Retrieval of Leaf Area Index and Tree Crown Parameters Using Terrestrial Laser Scanning

Leaf area index (LAI) is one of the most important biophysical parameters in biogeochemical cycles and global climate change studies. The abundant three-dimensional (3-D) information implicitly contained within the point cloud data (PCD) generated by a Terrestrial Laser Scanner (TLS) allows us to capture structural information of forest canopy directly from 3-D perspective. As a new emerging platform of Light Detection and Ranging (LiDAR) techniques, TLS has the potential to retrieve LAI directly from the 3-D PCD. The goals of this dissertation are to develop the methodologies and sampling protocols to nondestructively retrieve LAI directly from PCD, and examine the accuracy and precision of these measurements. This research only focused on the geometrical information (i.e. Cartesian coordinates (X, Y, Z)) implicitly contained within PCD. In general, I developed novel techniques to retrieve leaf orientation distribution, gap fraction, effective leaf area index (LAIe) from PCD using both 3-D and two-dimensional (2-D) based methods. Based on the geometrical computing, I developed methods to depict leaf orientation distribution. A method named "point cloud slicing" was developed in this research to characterize the 3-D canopy structure and LAI quantitatively. In addition, a TLS-based LAIe estimation method based on geometrical projection was developed. Finally, I developed the unique software titled "Toolkit for Terrestrial LiDAR" which allows the above-mentioned analysis. The methods and data capture protocol developed here compare well with desctuctive sampling of LAI and show our ability to potentially use the TLS to calibrate aerial laser scanning (ALS). The 3-D voxel data structure based point cloud slicing algorithm provides accurate structural information about the forest tree canopy and biophysical parameters such as LAI. In addition, the 2-D geometrical projection developed in this research allows us to retrieve LAIe from PCD regardless the light environment condition in forest stand. TLS-based LAI estimation will enhance our ability to provide 3-D structural information and calibration data for long-term ecological and global climate change studies.