Retrieval Of Clumping Index Of Forest Canopy Based On LiDAR Data

Forest leaves have many functions such as intercepting rainfall,absorbing light and gas,which is the main carrier of photosynthesis and other physiological functions.Clumping index is used to describe the spatial distribution of forest canopy leaves,and is necessary for quantitatively characterizing the non-random distribution of leaf elements in forest canopy.Meanwhile,it plays an important role in accurately expressing the distribution of forest canopy leaves and obtaining accurate LAI,etc.Therefore,extracting clumping index rapidly and accurately is of great significance for various physiological,climatic and biogeochemical studies.LiDAR can acquire three-dimensional information of canopy structure,which has been widely used in inversion of horizontal and vertical structure parameters.Ground-based and airborne LiDAR data have been successfully applied to canopy clumping index inversion.The ground-based LiDAR can get abundant point cloud to clearly reflect the distribution of forest canopy leaves,but the acquisition area is small.Airborne LiDAR can easily acquire point cloud data in large area,but the point cloud data is sparse.Taking the birch and larch in the greater hinggan mountains of Inner Mongolia as research objects,this paper proposed a new clumping index inversion method based on ground-based and airborne LiDAR data.The main contents of this paper are as follows:(1)Experiment Design and Data Processing:The three-dimensional point cloud data of the forest samples was acquired by Leica ScanStation C10 scanner.Digital hemisphere pictures of the forest sample were acquired by the digital camera equipped with a 180°wide-angle lens based on digital hemispherical photography technology.The pre-processing steps such as denoising and registration were performed on the terrestrial point cloud data.The binary classification was performed on the projected hemisphere pictures.The 3d point cloud data of forest quadrangle was obtained by the RIEGL lms-q680i scanner,and the airborne point cloud data was preprocessed by filtering,echo point classification and strip splicing.(2)Clumping index inversion of terrestrial LiDAR data:Considering the shortcomings of the voxelization method,this paper proposes a model to divide the hemisphere surface into regions.A new clumping index inversion method for forest canopy was developed based on high-resolution terrestrial LiDAR point cloud data.This method consist of operations such as converting rectangular coordinates into spatial polar coordinates and hemispheric surface projection.Using digital hemisphere photography technology,the canopy clumping index of the digital hemisphere image in the same research area is calculated,and the results are correlated.The R~2 between the two methods is about 0.75.The results show that the new method proposed in this paper has a higher accuracy than the digital hemispheric photograph method.Compared with the existing voxel method,the new method not only improves the accuracy and computational efficiency,but also enhances the processing ability of larger space and denser data.(3)Airborne LiDAR data clumping index inversion:Aiming at the shortcomings of the model,the improved model for calculating the gap fraction is proposed according to Bill's law.The gap fraction is calculated using the existing first-echo number ratio model,all-echo number ratio model and echo number intensity ratio model.Then,the results are inverted according to the logarithmic gap averaging method.Finally,the index of canopy clumping index was mapped.The results of four kinds of model calculation and the terrestrial data inversion were analyzed.The results show that the modified echo intensity ratio method proposed in this paper is most suitable for inversion of large range clumping index,the R~2 between the two methods is about 0.6,and the clumping index is related to topography and affected by slope and orientation.