毛竹雨雪冰冻灾害主要影响因子的回归分析

应用逐步引入剔除法对毛竹受雨雪冰冻灾害程度的海拔、坡度、坡位、坡向、土壤深度、经营管理、竹林密度等因子与折断率、倒伏率、受损率分别进行多元线性回归分析,经检验F>F0.01,P<0.01,偏回归系数(bi)均为显著或极显著,说明回归方程均有极显著意义。通过标准偏回归系数(b′i)检验自变量的主次关系,表明影响折断率的主导因子及各因子的主次关系为海拔>土壤深度>坡位>坡度,海拔、坡度与折断率呈正相关,坡位、土壤深度与折断率呈负相关;影响倒伏率的主导因子及各因子的主次关系为竹林密度>坡度>经营管理>土壤深度,坡度、经营管理与倒伏率呈正相关,竹林密度、土壤深度与倒伏率呈负相关;影响受损率的主导因子及各因子的主次关系为土壤深度>坡度>海拔>竹林密度>经营管理,海拔、坡度、经营管理与受损率呈正相关,竹林密度、土壤深度与受损率呈负相关。说明回归模型可用于毛竹折断率、倒伏率和受损率预测。

Regression analysis of impact factors related to snow disaster of Phyllostachys pubescens

At the beginning of 2008, southern China suffered severe winter freezing storms, which caused heavy losses of bamboo forest, especially Phyllostachys pubscens. In order to be aware of form the impact factors related to the disaster, stepwise selection method was used to analyze the relationship between factors such as altitude, slope, slope position exposure, soil thickness, management intensity and stand density, and damage degree such as break ratio, lodge ratio and damage ratio. The results showed that multiple regression equations were significant (F>F0.01, P<0.01). The degree sequence of leading factors related to break ratio was altitude>soil thickness>slope position>slope, with altitude and slope being positively related while stand density and soil thickness negatively. The sequence related to lodge ratio was stand density>slope>management intensity>soil thickness, with slope and management intensity being positively related while stand density and soil thickness negatively. As to damage ratio, the sequence was soil thickness>slope>altitude>stand density>management intensity, with altitude, slope and management intensity being positively related while stand density and soil thickness negatively.