基于回归模型的广安区古树名木资源特征研究

The Resource Characteristics of Old Trees in Guangan District based on Regression Model

在实地调查、测量、走访和查阅大量文献的基础上,运用统计学、回归模型对广安区121株古树名木资源特征、树龄与树高、冠幅、胸径单因素结构、一元二次回归和多元线性回归模型进行分析。结果表明,研究区内古树名木隶属11科12属15种。各个树种之间数量差异上异常明显,主要以桑科中的黄桷树、小叶榕优势种群突出;单因素结构模型分析发现古树名木树龄、树高、冠幅呈拉普拉斯模型、高斯模型、高斯模型、对数模型分布;一元二次回归效果好,精度高,树龄与树高、胸径、冠幅呈同步增加趋势;树龄、树高、冠幅、胸径间符合多重线性回归模型,且整体回归模拟较强。通过树高、冠幅、树龄和胸径多元回归估算树龄能为以后古树名木普查树龄(主要系黄桷树、银杏树)的估算提供一种切实可行的方法。

Based on the field survey and literature review,this paper analyzed 121 old trees in Guangan district by using statistical and regression models. The results showed that there were 11 families,12 genera and 15 species in the study area. The difference in the number of different tree species was very obvious,and the dominant population were Ficus virens Ai,Ficus benjamina mulberry. Single factor structure model analysis revealed that the tree age,tree height,crown width were Laplace model,Gauss model,Gauss model,and logarithmic model distribution. The effect of the quadratic regression was good and the precision was high. The tree age,tree height,crown width and DBH increased synchronously. The tree age,tree height,crown width and DBH were in line with multiple linear regression models,and overall regression simulation is strong. The estimation of tree age by multiple regression of tree height,crown,tree age and DBH can provide a practical method for the estimation of tree age( mainly Ficus virens Ai and Ginkgo).