Welcome to Bulletin of Botanical Research! Today is Share:

Bulletin of Botanical Research ›› 2015, Vol. 35 ›› Issue (6): 923-928.doi: 10.7525/j.issn.1673-5102.2015.06.021

Previous Articles     Next Articles

Effects of Different Rare Species Treatments on Two-way Indicator Species Analysis

KANG Yan-Ling;ZHANG Qin-Di*;DUAN Xiao-Mei;LI Ting-Ting;BI Run-Cheng   

  1. College of Life Sciences,Shanxi Normal University,Linfen 041004
  • Online:2015-11-20 Published:2016-01-18
  • Supported by:
     

Abstract: For comparing the impact of different treatment methods on TWINSPAN, we studied two different methods of eliminating rare species and untreated with the plant communities and environmental survey data of Xiaolongmen Forest Farm. Match coefficient and Davies-Bouldin index(DBI) were taken to compare the result of two different TWINSPAN classification for ensuring the best classification division. Before and after eliminating accidental species, the TWINSPAN classification result among the same classification principles of termination was, respectively, divided into 12 and 11 associations. With the combination coefficient r, the low grade of TWINSPAN classification results of before and after eliminating rare species were not consistent, with the increase of classification level, the ground adhesion coefficient was higher and higher, and the rare species greatly influenced the result of the low grade of TWINSPAN classification. DBI index as the division criterion was introduced to obtain the optimum TWINSPAN classification results. Although the best classification of grade was different before and after eliminate rare species, but with higher classification results of alignment the degree of agreement between two ones was higher. Therefore, in the application of TWINSPAN classification, when introducing DBI index to determine the best classification, the low classification results of rare species should be deleted.

Key words: TWINSPAN classification, rare species, DBI, contingency tables, Xiaolongmen Forest Farm

CLC Number: