APPLICATION OF MULTIVARIATE STATISTICAL ANALYSIS METHODS TO THE STUDY ON THE HUMAN IMPACTS ON THE WATER QUALITY OF THE DU TRIBUTARY IN THE CAU RIVER BASIN

Abstract

Monitoring program is crucial aspect of the river management. However it is more important to make interpretation of the monitoring data in order draw an appropriate management measures for restoration and improvement of river water quality. Within the context of this study, we posed two problems: (1) identification of the factors that influence the water quality of the river and (2) clustering monitoring sites according to ecological health. Resolving these two problems necessitated the application of multivariate statistical analysis methods, including parametric and non-parametric correlations, canonical correlation analysis and hierarchical cluster analysis using Bray-Curtis distance. These methods thus are being applied to analysis on human impacts on ecological health of the river based on distribution of macroinvertebrate communities. As for the factors influencing the river health, multivariate statistical analysis exposed the influence of the metal concentration (total Fe), dissolved oxygen, and to a lesser degree, the nutrient enrichment