Individual Tree Structural Parameter Extraction and Volume Table Creation Based on Near-Field LiDAR Data: A Case Study in a Subtropical Planted Forest.

Sha Gao,Lin Cao,Zhengnan Zhang

doi:10.3390/s21238162

Abstract

Individual tree structural parameters are vital for precision silviculture in planted forests. This study used near-field LiDAR (light detection and ranging) data (i.e., unmanned aerial vehicle laser scanning (ULS) and ground backpack laser scanning (BLS)) to extract individual tree structural parameters and fit volume models in subtropical planted forests in southeastern China. To do this, firstly, the tree height was acquired from ULS data and the diameter at breast height (DBH) was acquired from BLS data by using individual tree segmentation algorithms. Secondly, point clouds of the complete forest canopy were obtained through the combination of ULS and BLS data. Finally, five tree taper models were fitted using the LiDAR-extracted structural parameters of each tree, and then the optimal taper model was selected. Moreover, standard volume models were used to calculate the stand volume; then, standing timber volume tables were created for dawn redwood and poplar. The extraction of individual tree structural parameters exhibited good performance. The volume model had a good performance in calculating the standing volume for dawn redwood and poplar. Our results demonstrate that near-field LiDAR has a strong capability of extracting tree structural parameters and creating volume tables for subtropical planted forests.

Highlights

Published: 6 December 2021A volume table is critical for forest resource inventory, forest growth monitoring, forest volume statistics, etc. [1]
Our results demonstrate that near-field light detection and ranging (LiDAR) has a strong capability of extracting tree structural parameters and creating volume tables for subtropical planted forests
Point cloud data can directly reflect the structural information of a forest canopy, but traditional individual tree segmentation based on remote sensing images cannot obtain the full structural information of a forest, so the segmentation accuracy of individual tree segmentation using LiDAR point cloud data is usually higher than that of traditional remote sensing image extraction

Summary

Introduction

Published: 6 December 2021A volume table is critical for forest resource inventory, forest growth monitoring, forest volume statistics, etc. [1]. The accurate assessment of volume and the growth of planted forests are of great importance in maintaining regional and global forest ecosystems, as well as in scientific decision making concerning planation cultivation [3,4]. As the forestry sector standard in China, the local volume table (i.e., one-way volume table), which allows the estimation of the stem volume from the diameter at breast height (DBH), originated in the 1970s. It is an urgent problem to redevelop the standard volume model with a high precision and a strong applicability and to compile a standard volume table (i.e., two-way volume table, requiring both DBH and tree height) for the precision of cultivation of planted forests depending on the different requirements for forestry production and forest cultivation in different areas. The forest volume is an important parameter for assessing forest health, Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations

Objectives

Methods

Results

Discussion

Conclusion