According to the field data of surface water quality from Beijing Water Authority, this paper demonstrates a case study on how to utilize principle factor analysis and hierarchical cluster analysis to extract a limited number of principal factors that can best describe the original data and to identify the patterns of surface water quality pollution. 10 auto-monitoring sites dispense in Beijing are selected as study objects, and according to actual monitoring data of comprehensive database from Beijing Water Authority, 9 indicators including: WT, PH, NO 3 N, NH 4 N, DO, CNDR, TRB, DLS and CHLA are selected for principle factor analysis and hierarchical cluster analysis. Based on the monitoring data during 2010, principle factor analysis is utilized to reflect those chemical data with the greatest correlation, and the results identify four principal factors representing 92.432% of cumulative variance (or total information). By utilizing principle factor analysis, thermal pollution factor, nitrate pollution factor, plankton pollution factor and ammonia nitrogen pollution factor reasonably interpret the main factors of surface water quality pollution. Based on factors' scores, the level of the comprehensive water quality pollution about the 10 current auto-monitoring sites is obtained, by which condition of surface water quality pollution is sorted. The results show that Gaobeidian and Sanjiadian are heavily polluted, which are tally with Beijing Water Resources Bulletin of 2010. Further more, in support of the results obtained by principle factor analysis, 5 clusters are assigned by HCA to evaluate the similarities of water quality among the 10 automonitoring sites. In allusion to the different patterns of surface water quality pollution and the influencing factors, suitable microscopic measures can be proposed for different areas, which can provide a base for macroscopic planning of the city.
Keywords-surface water; water quality assessment; principle factor analysis; hierarchical cluster analysis; beijingI.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.