Application of Multivariate Statistics Simplifying the Indicators of Lake Ecosystem Health Assessment
Several different definitions and various indicators have been proposed and discussed in the ecosystem health assessment. However, due to the lack of quantitative criteria and the correlations among the indicators, evaluating the ecosystem health by these indices may result in some deviation. Aimed at this problem, the principal component analyses (PCA) and the principal factor analyses (PFA) technique were applied to evaluate the effectiveness of the health indicators. The objective was to identify health parameters that are important and reveal more information in assessing lake ecosystems health status. Twenty-eight variable including physical, chemical, biological, system-level aspects were sampled from thirty-five lakes in Wuhan city, P. R. China for the purpose in this study. The results showed that these techniques play excellent performance in select and simplify index for the assessment, in addition, the. results also showed that three kinds of indices occupy the relatively important position in the assessment; 1) index of anthropogenetic stress; 2) the heavy metal parameters; 3) eutrophication index. The Shannon-weaver diversity index, temperature and specific exergy had the lowest correlation coefficients values (<0.6 with all factors), which could indicate that these parameters were less important in reveal lake ecosystem health variations. In addition, an assessing model (IHI) for lake ecosystems by statistical method was proposed.
Ecology lake ecosystem ecosystem health assessment principal component analysis principal factor analysis
LU Jin WU Hongjuan LI Yuan CHEN Mengqiu
School of Environmental Science and Engineering, Huazhong University of Science and Technology, Wuhan 430074, Hubei, China
国际会议
2007年中国科协科技年会(The Annual Conference of the China Association for Science and Technology)
武汉
英文
157-162
2007-09-08（万方平台首次上网日期，不代表论文的发表时间）