林业科学2026，Vol.62Issue(7)：27-37,11.DOI:10.11707/j.1001-7488.LYKX20250603
基于UAV-LiDAR数据的长白落叶松多维偏冠的邻体效应
Neighbor Effects on Multi-Dimensional Crown Asymmetry in Larix olgensis Based on UAV-LiDAR Data
摘要
Abstract
[Objective]Traditional forest inventory methods are limited in accurately characterizing asymmetric crown growth.This study aims to develop multi-dimensional crown asymmetry indices using UAV-LiDAR point cloud data and to evaluate the effects of neighbor interactions on the degree of crown asymmetry.[Method]The study was conducted in a Larix olgensis plantation in Mengjiagang Forest Farm,Jiamusi City,Heilongjiang Province.An UAV-LiDAR platform was used to acquire point cloud data.Following individual tree segmentation,crown parameters were extracted to construct asymmetry indices from three dimensions to quantify crown asymmetry.The one-dimensional crown asymmetry index(CAI1)was defined as the angle between the treetop and the centroid of the 3D convex hull of the crown.The two-dimensional index(CAI2)represented the deviation of the crown projection area from a standard circle.The three-dimensional index(CAI3)quantified the deviation of the actual 3D crown volume from an assumed ideal symmetrical geometry of tree crown.Subsequently,factors related to crown size,spatial structure,and neighbor competition were extracted.Correlation analysis and linear mixed-effects models(LMM)were employed to assess the influence of different neighbor characteristics on multi-dimensional crown asymmetry.[Result]The results showed that crown size factors were all significantly negatively correlated with crown asymmetry indices,meaning that the degree of asymmetry decreased as crown size increased.Among the spatial structure indices,uniform angle index(W)and crowding(C)showed no significant effect on asymmetry(P>0.05).However,tree height and crown neighborhood comparison,as well as opening degree,were significantly correlated with all three asymmetry indices(P<0.01),directly driving crown shape development.Neighbor competition indices were very significantly positively correlated with CAI2 and CAI3,indicating that crown asymmetry was intensified with increasing competitive pressure,revealing the critical role of neighbor competition in driving crown asymmetry.Trees tend to expand their canopy preferentially towards more open spaces to avoid lateral competitive pressure from neighbors,leading to asymmetric crown development.Compared to CAI2 and CAI3,CAI1 values were more concentrated(0-0.12)and showed generally weaker correlations with tree size,spatial structure,and neighbor competition.LMM showed that crown size,spatial structure,and neighbor competition factors explained up to 65%of the variance in CAI3,substantially higher than the 19%explained for CAI2.Analysis of standardized fixed effect coefficients revealed that variation in crown asymmetry was primarily driven by neighbor competition,with variation in CAI3 resulting from the combined effects of both crown size and neighbor competition.[Conclusion]Competition of trees for growing space and light resources is a key factor leading to asymmetric crown development.Neighbor competition plays a dominant role in driving variation in the CAI2 and CAI3 asymmetry indices.Correlation analyses between spatial structure factors and asymmetry indices indicates that,compared to the horizontal distribution pattern of trees,their size dominance relationships and vertical spatial competition have a more direct effect on crown asymmetry.The differential explanatory power of LMM for CAI2 and CAI3 variation confirms that a three-dimensional spatial perspective captures localized crown shape adjustments more effectively,revealing that the essence of tree interactions is competition for spatial niches.关键词
偏冠指数/树冠结构/林木空间结构/邻体竞争/无人机激光雷达Key words
crown asymmetry index/crown structure/tree spatial structure/neighbor competition/unmanned aerial vehicle分类
农业科技引用本文复制引用
张诺,郝元朔,董利虎,赵颖慧,李凤日..基于UAV-LiDAR数据的长白落叶松多维偏冠的邻体效应[J].林业科学,2026,62(7):27-37,11.基金项目
国家重点研发计划课题(2023YFD2200802) （2023YFD2200802）
国家自然科学基金青年科学基金项目(32301580). （32301580）
