Study On The Estimation Method Of Forest Parameters Using LiDAR And SAR

Carbon, water, and energy is the ecological exchanged processes between forest and atmosphere They are influenced by forest canopy and stand structure. Therefore, exploring the technique and method of quantizing canopy structure and stand structure to monitor the change of forest structural parameters is very significant. Forest parameters(e.g, the distribution of forest, height, DBH, biomass, LAI and forest stock volume) is the indispensable part of forest survey, the traditional survey method is difficult to survey large area, also the visible spectral remote sensing have made a huge contribution on forest survey, it has disadvantages in detecting in the vertical structure of forest. The main purpose of this dissertation is to investigate the the potential and feasibility of deriving forest parameters from LiDAR carried on satellite, Airborne and SAR in different scales. Specifically, this dissertation mainly conducted some research as follows:1. Forest height model, biomass model and LAI model have been constructed,based on data preprocessing, filtering, and wave parameters extraction. The parameters of height model are canopy height extent (Lvh), canopy valid extent (Lv) and ground valid extent (Lg), R2is0.86. the key parameters of biomass model are entire wave energy (Ev) and canopy height extent (Lvh), R2is0.91. The parameter of LAI is the ratio of ground wave energy and entire wave energy, R2is0.84.2. This dissertation summarize the application and method of synthetic aperture radar (SAR) in forest survey. With the analysis of relationship between back scattering coefficient and forest stock volume, a forest stock volume model was given, and R2is0.512.3.This paper summarizes the work principle and application of LiDAR in forest survey, then discussed the efficiency of the LIDAR data Processing. In order to deriving the forest canopy, Canopy Height Model (CHM), the subtraction of digital elevation model (DEM) and digital surface model (DSM), were made, and the DEM and DSM had been interpolated from laser points cloud.4. A tree extracting algorithm is presented from CHM, the algorithm based on segmentation of CHM and the peaks identification. Individual tree parameters (e.g, tree height, crown breadth, DBH, biomass) were extracted from the CHM. The CHM percentage was used to extracted forest stand parameters.5. CHM was introduced into the classification of land cover, the height characteristic of CHM differentiate the land covers which has similar spectral features but different height (e.g, shrub and alpine meadow), classification accuracy up to90.25, the Kappa coefficient is0.86.This research mainly focus on the forest parameters estimation (e.g, the distribution of forest, height, DBH, biomass, LAI, forest stock volume and so on), and specifically, the parameters related to vertical structure. The satellite-based LiDAR provide the possibility of estimation of tree height, biomass and LAI in big area and even global. The SAR and Airborne LiDAR provide the possibility of precise estimation of forest parameters.