Phylogenetic Structure, Numerical Classification And Ordination Of Mixed Broadleaf-conifer Forest Communities Along The Elevation Gradient In Meill Snow Mountain

Meili Snow Mountain is located in eastern slope of Himalayas Mountains and northwest of Yunnan Province, at the convergence of Yunnan and Tibet. Meili Snow Mountain is one of the best places for vegetation ecology research. This study focused on the mechanism of species coexistence in mixed broadleaf-conifer forest along elevation gradient in Meili Snow Mountain. The elevation in research area ranged from2750m to3300m. We collected75arbor sample plots and13herbal sample plots. We used two phylogenetic indexes NRI (net relatedness index) and NTI (nearest taxon index) to detect the community phylogenetic structure in mixed broadleaf-conifer forest. An on-line software named Phylomatic(Version3) was used to build a phylogenetic tree. And Phylocom software was used to calculate phylogenetic indexes. We also analyzed the quadrats data using various methods of community classification and ordination, average linking clustering(UPGMA), principal component analysis(PCA) and correspondence analysis (CA). Our results were as follows:1. Results of communities phylogenetic structure(1) Niche conservatism existed for species of mixed broadleaf-conifer forest communities in this area:Species phylogenetic diversity of mixed broadleaf-conifer forest community had a significant positive correlation with species pool (P<0.001).(2) In lower elevation area (2750m-3100m), in mixed broadleaf-conifer forest communities, arbor species phylogenetic structure was characterized by phylogenetic clustered. It seems that habitat filtering was stronger than competition exclusion in this area. Disturbance (such as grazing) might have important impact on the clustered communities phylogenetic structure. In higher elevation area (310Om-3300m), communities were phylogenetically over dispersed. It is notable that this area was suitable for arbor species growth, and competition exclusion might be stronger than habitat filtering in this area. With increasing elevation, herbal species showed no clear pattern. Compared to elevation, soil and slope might have a greater influence on herbal species. 2. Results of communities numerical classification and ordination(1) Arbor species richness had a significant negative correlation with elevation (P<0.001). Arbor species Shannon-Wiener diversity index also had a significant negative correlation with elevation (P<0.05). The reason could be that environment, with increasing elevation, was progressively deteriorating. This lead to the decreased species richness and Shannon-Wiener diversity index.(2) The75quadrats were grouped into6clusters by UPGMA method. Each cluster was located in different elevation. The reason could be that different species existed in different elevation. Pattern of the6clustering groups were almost the same in ordination diagrams of CA and PCA. According to PCA ordination, distribution of Acer pectinatum had a positive association with Betula utilis, and distribution of the two species had a negative association with Pseudotsuga forrestii. The ordination result of CA and PCA indicates that altitude had great influences on mixed broadleaf-conifer forest communities.