Research on Tree Carbon Storage Estimation Method based on Point Cloud and Voxel Modeling

Abstract

Since global climate change has become more pronounced, accurate estimation of forest carbon stocks has become a vital part of carbon cycle research and carbon neutrality strategies. However, it is well known that traditional methods for estimating tree carbon stocks rely mainly on empirical models and manual measurements, both of which have inherent problems of low accuracy and low efficiency. Therefore, recent developments in 3D laser scanning offer promising solutions.since LiDAR technology has provided new and effective methods for precise forest structure measurement, this paper makes a very clear and logical use of high-density point cloud data acquired by the Oslight Intelligent R8 backpack SLAM laser scanning system for vegetation on the campus of Liaoning University of Science and Technology, and then applies a voxelization modeling approach to compute tree volumes and estimate carbon stocks.Using Lidar360 software for individual tree segmentation, parameter extraction, and volume calculation, the results were systematically compared with those from MATLAB voxelization algorithms, and it was clearly and convincingly shown that a voxel resolution of 5 cm gives the most accurate tree volume measurements. Therefore, the paper naturally leads into discussing the advantages of voxelization algorithms in representing complex tree structures while maintaining computational efficiency, making it an excellent foundation for a high-precision, automated method of urban forest carbon sink monitoring. It also provides a solid technical reference for high-precision carbon stock estimation from point cloud data.