森林工程2018，Vol.34Issue(2)：35-39,5.
森林生物量的空间自相关性研究
Study on Spatial Autocorrelation of Forest Biomass
摘要
Abstract
Forest biomass is the basis of material circulation and energy flow of forest ecosystem, which can measure forest productivity well. It is of great significance to study the spatial distribution and variation of forest biomass and to reveal and to reveal the law of surface space change. One of the main areas of this research was the region of Mao'er Mountain forest, and the global autocorrelation analysis and local autocorrelation analysis were carried out on the biomass of Mao'er Mountain forest region, according to the survey data of the fixed sample plots in Mao'er Mountain region in 2004. Then, R software was used to establish multiple linear regression model of biomass, geographical factors and biological factors. GWR4.0 software was used to establish the geographical weighted regression model with the number of plants per hectare, the average diameter at breast height and the elevation as explanatory variables. The results showed that the forest biomass in the Mao'er Mountain area had positive spatial autocorrelation. In this study area, the AIC value of Geographical Weighted Regression Model was 90 lower than the multi-regression model, R square value and adjusted R square value were both increased, and the GWR model had higher fitting accuracy.关键词
生物量/空间自相关性/多元线性回归模型/地理加权回归模型Key words
Biomass/spatial autocorrelation/multiple linear regression model/geographical weighted regression model分类
农业科技引用本文复制引用
王维芳,董薪明,董小枫,吕丹阳,苏婷婷,郑安然..森林生物量的空间自相关性研究[J].森林工程,2018,34(2):35-39,5.基金项目
大学生创新训练计划项目号201610225155 （）
