Fuzzy clustering analysis of insect community structure in a Zanthoxylum bungeanum garden
Author of the article:ZHANG Xiao-Ming1** LI Qiang1*** CHEN Guo-Hua1*** SHI An-Xian2 SONG Jia-Xiong2
Author's Workplace： 1. College of Plant Protection, Yunnan Agricultural University, Kunming 650201, China; 2. Plant Protection and Quarantine Station of Zhaotong, Zhaotong, Yunnnan 657000, China
Key Words： Zanthoxylum bungeanum, insect community, structure temporal dynamic, fuzzy clustering
Abstract： [Objectives] To provide a scientific basis for the ecological regulation of pest insects and build a three-dimensional agricultural system in a Zanthoxylum bungeanum garden. The insect community and its structure were sampled through systematic investigation, its spatiotemporal dynamics analyzed and the relationship between each sub-community evaluated. [Methods] The five points sampling method was used to record the number of insect species on branches and trunks in Z. bungeanum trees and a combination of sugar-acetic acid-ethanol traps, yellow sticky traps, yellow-drip plate traps, visual inspection and sweep sampling were also used to sample the insect community. Hierarchical clustering was used to classify the temporal dynamics of various community indices. The basic characteristics of the insect community were analyzed using basic features of the community characteristics index. [Results] 359 species of insect (including spiders) belonging to 19 orders, 148 families were collected. Based on the principles of community ecology and the method of fuzzy clustering, the results show that the Hemiptera were the most abundant and spiders the second most abundant taxa, comprising 0.3036 and 0.2033, of all species collected, respectively. 24 investigations were conducted on the structural features of the insect community, and a cluster analysis was conducted on its temporal dynamics．The total community could be clustered into 9 temporal clusters (T=0.98), i.e., 20 March, 20 May, 30 March, 28 April, 10 April, 10 October; 20 October, 30 October, 12 November, and other months. The natural enemy sub-communities could be clustered into 7 temporal clusters (T=0.95), i.e., 20 March, March 30, 30 May, 12 November; 20 April, 28 April, 10 April, 30 October, and other months. The pest-neutral sub-communities could be clustered into 9 temporal clusters (T=0.97), i.e., 20 March, 30 March, 20 May, 20 June, 30 September, 10 October, 10 June, 30 June, 20 October, 30 October, 12 November, and other months. [Conclusion] Cluster analysis partly reflected seasonal differences in the total community and sub-communities, and the temporal overlaps between clusters reflect the complexity of the community structure. The complexity of cluster results reflects the period when non-biological factors had the largest impact on the community.