Chapter 10 - LiDAR-Based Three-Dimensional Radiative Transfer Models and Applications

Abstract

Modeling radiative transfer (RT) in heterogeneous forests is essential to understanding their biophysical processes and retrieving information from remote sensing data. Although a range of three-dimensional (3-D) RT models have been reported, it is still challenging to obtain realistic 3-D scenes and input parameters using traditional field surveys or optical imagery. Light detection and ranging (LiDAR) can capture highly detailed point clouds of forests to provide 3-D scenes and canopy structural information for parameterizing 3-D RT models. This chapter introduces the most common 3-D RT models. We then focus on the voxel-based RT (VBRT) model, which uses octree and high-resolution voxels rather than coarse-resolution turbid medium voxels to represent forest scenes. First, we describe 3-D scene construction and parameterization using LiDAR point cloud data in detail. We then examine and compare simulated bidirectional reflectance factor and orthogonal photographs using VBRT and two additional 3-D RT models. Finally, we summarize the applications of 3-D RT models combined with other RT or process models to simulate the directional reflectance spectra of forests and ecological processes of forest ecosystems, such as gas exchange and solar-induced chlorophyll fluorescence.