Tree Richness and Climatic Dominance in Different Districts of Beijing City, China

Abstract

Nonparametric richness estimators, generalized linear models, canonical correspondence analysis and indicator species analysis were applied for revealing the tree diversity and structure in the districts of Beijing City, China. The results indicated that Mentougou has the highest richness of tree species, occupying 75.8% of total species concerned. Based on the non-parametric richness estimators, the tree richness in Beijing area was estimated around 71-79. From the dendrogram of Wards minimum variance algorithm, the studied districts can be grouped into three main groups, the first two of which have representative species as selected by indicator species analysis. Group III (including districts Yanqing and Pinggu) does not have representative species, implying that this areal group is not valid if more species distribution is added in the cluster analysis. Environmental correlation based on generalized linear models found that the current tree richness in Beijing City is significantly associated to the following factors: minimum temperature of coldest month, mean temperature of driest quarter and precipitation of driest quarter, which can take account of 95.8% of the total deviance in species richness. Based on complementary principle, the most valuable districts for tree conservation in Beijing were Mentougou, Miyunxian, Haidianqu and Yanqingxian respectively.