基于无人机激光雷达的刺槐人工林森林健康指标的构建

Establishment of forest health indicators of Robinia pseudoacacia plantation based on UAV LiDAR

利用无人机激光雷达LiDAR(Light detecting and ranging)数据,提取能反映植被垂直和水平结构变化的LiDAR特征变量,通过相关性分析和层次聚类方法构建森林健康指标来识别黄河三角洲刺槐人工林的健康状况。结果表明,森林健康指标由LAD_(cv)(叶面积密度的变异系数)、weibull_α(Weibull密度函数的尺度变量)、H_(99)(高度百分位数)和VCI(垂直复杂度)构成;利用森林健康指标进行刺槐林的健康等级判断可以得到较理想的结果(总精度为86.7%,Kappa系数为0.79),证实了激光雷达技术在判断森林健康状况方面的潜能。

LiDAR(Light detection and ranging)data of UAV was used to extract LiDAR characteristic variables that can reflect the changes of vegetation vertical and horizontal structure.Forest health indicators were constructed by correlation analysis and hierarchical clustering method to identify the health status of robinia pseudoacacia plantation in the Yellow river delta.The results showed that,the forest health indicators were composed of LAD_(cv)(the coefficient of variation of leaf area density),weibull_α(the scale parameter of the Weibull density function),H_(99)(the percentile height of 99 th)and VCI(Vertical complexity index);Using forest health indicators to judge the health status of Robinia pseudoacacia forest could get an ideal result(total accuracy of 86.7%,Kappa coefficient of 0.79),which confirmed the potential of LiDAR technology in forest health assessment.