1. Hulunber Grassland Ecosystem Research Station, Beijing 100081, China;2. Key Laboratory of Resources Remote-Sensing and Digital Agriculture of Ministry of Agriculture, Institute of Natural Resources and Regional Planning, Chinese Academy of Agricultural Sciences, Beijing 100081, China;3. International Institute for Earth System Science of Nanjing University, Nanjing 210093, China
Abstract:Numerical characteristics such as species composition and its relationship with environment are studied based on field survey and TWINSPAN and CANOCO software.The dominant families are Compositae,Gramineae,Leguminosae,Liliaceae,and Rosaceae,which representing 13.4%,11.3%,10.3% and 8.2%,8.2%,respectively,of the collected 96 species from 66 genera and 25 families,and perennial herbages are most abundant.Communities including 96 samples are divided into 13 groups by TWINSPAN according to the community characteristics.The result indicated that spatial pattern of the associations were mainly affected by sample position,soil nutrition,soil water and pH.DCA analysis shows that the ordination diagram of dominant species was similar to that of the samples,and ordination results couple well with classification outputs,which implicate that variation of community and species distribution are determined by change of environmental factors.