【目的】利用地基激光雷达技术(Terrestrial Laser Scanning,TLS)获取树木干形特征的可行性并对红松进行种源评价选择,为红松优良种源的综合评价提供理论依据。【方法】以黑龙江省帽儿山种源试验林中的26个红松种源为试验材料,首先借助TLS扫描样地,通过LiDAR360软件对数据进行预处理,对TLS提取的树高、胸径及不同高度处直径等参数进行评价;计算4种干形参数(胸高形数、胸高形率、实验形数、高径比)并利用TLS提取的不同高度处直径构建基于种源效应的削度方程,描述不同种源树干干形变化。【结果】地基激光雷达对单木参数提取都达到了较高的精度,其中胸径的提取效果(决定系数R²=0.979 6)优于树高(R²=0.811 4)和材积(R²=0.911 9);随着树干高度的增加相对高度处直径提取的R²逐渐增大,相对高度为0.08h(h为树高)时R²达到最大值(R²=0.984 0),相对树高大于0.08h后,相对高度处直径提取的R²逐渐减小;对26个种源红松干形指标的方差分析结果表明,各指标在种源间均达到了显著差异(P<0.05);将种源变量引入基础模型,模型的拟合精度提高,以重复作为随机效应构建非线性混合模型,模型精度进一步提高;通过绘制不同种源干形图,各种源干形变化趋势基本一致,但生长速度存在差异。【结论】TLS对单木参数提取精度较高;各干形指标在种源间均达到显著差异,具有较好的选择潜力;根据干形指标和基于种源效应的最优削度方程筛选出优良种源,可为红松的遗传改良和推广使用提供理论依据。

【Objective】This research aims to explore the feasibility of utilizing terrestrial laser scanning (TLS) technology for acquiring stem form characteristics of trees and to evaluate and select elit of Korean pine (Pinus koraiensis), providing a theoretical basis for the comprehensive evaluation of superior provenances.【Method】Using 26 P. koraiensis provenances from the provenance test forest in Maoershan, Heilongjiang Province, as experimental materials, TLS was employed to scan sample plots. Data preprocessing was conducted using LiDAR360 software, and parameters such as tree height, diameter at breast height (DBH), and diameters at different heights extracted by TLS were evaluated. Four stem form parameters (breast-height form factor, breast-height form quotient, experimental form factor, and height-to-diameter ratio) were calculated. Additionally, taper equations based on provenance effects were constructed using diameters at different heights extracted by TLS to describe stem form variations among different provenances.【Result】TLS achieved high accuracy in extracting individual tree parameters. The extraction accuracy for DBH (coefficient of determination R²= 0.979 6) was superior to that for tree height (R²=0.811 4) and volume (R²=0.911 9). As the stem height increased, the R² for diameters at relative heights gradually increased, reaching its peak at 0.08h(R²=0.984 0, where h represents tree height). Beyond 0.08h, the R² for diameters at relative heights gradually decreased. Variance analysis of stem form indicators for the 26 P. koraiensis provenances showed that all indicators exhibited significant differences among provenances (P<0.05). Introducing provenance variables into the base model improved the fitting accuracy, and constructing a nonlinear mixed model with replication as a random effect further enhanced model precision. By plotting stem form graphs for different provenances, it was observed that the trends in stem form variation were generally consistent across provenances, but growth rates differed.【Conclusion】TLS demonstrated high accuracy in extracting individual tree parameters. All stem form indicators showed significant differences among provenances, indicating good potential for selection. Based on stem form indicators and the optimal taper equation incorporating provenance effects, a group of superior provenances was identified, providing a foundation for genetic improvement and widespread utilization of P. koraiensis.