Abstrait

The information mining of biological monitoring data in a heavily polluted river: The comparison between shannon-wiener diversity index and the multivariate analysis

Feng Li, Xiang-Yun Zeng, Xiao-Lin Long, Yan-Yan Lang, Yan-Mao Wen

The information mining of biological monitoring data is important for environment monitoring and accessment. Although Shannon-Wiener diversity index (SWI) has beenwidely used to explain the results of aquatic biologicalmonitoring previously, it runs into problems in heavily polluted rivers. In this paper, a representative heavily polluted river has been selected, and the samples of sediment, pore water, phytoplankton, zooplankton, and zoobenthoswere collected and analyzed, with a view to providing theoretical basis for biological data analysis in heavy polluted area. SWI, themultivariate analysis (combined by twomultivariate analysis methods: cluster and Non-matricMulti-dimentional Scaling analysis) were used to analyze biological data of phytoplankton, zooplankton, and zoobenthos, with the results of the physical and chemical monitoring and assessment as reference. The results show that the results of SWI cannot effectively reflect the difference of pollution status of various stations in the heavily polluted river; despite the presence of some problems, multivariate analysismethod ismore suitable than SWI as far as information mining of biologicalmonitoring in the heavily polluted river is concerned.