Using Multivariate Statistical Techniques to Examine the Spatial Variability of Physicochemical Elements and Water Quality in Hong Kong's Rivers

Abstract

Hong Kong is a metropolitan city located at the South East coast of China. Owing to the geographical location, landuse pattern and rapid economic growth or development, the rivers in Hong Kong are influenced by natural, marine and anthropogenic factors. These are reflected in the spatial variability of physicochemical elements in different watercourses. The 6-year river data water quality set measured by the Hong Kong Environmental Protection Department (HKEPD) at 82 monitoring sites are used to study the characteristics and distribution of physicochemical variables in local rivers. Some 27 determinands of surface water quality (pH, water temperature, flow, Salinity, Total Solid, Conductivity, Chloride, TOC, BODs, COD, DO, DO%, NH4, N02, NO.?, TKN, TP, P04, Al, Fe, Zn, Mn, Pb, Cd, CI; Cu, Ni) are examined with two multivariate statistical techniques: Cluster analysis and Factor analysis. The results indicate that nutrients, aggregate organics and heavy metal elements have greater spatial variation than physical parameters which are less influenced by anthropogenic factors. In addition, these elements can be separated into two clusters reflecting marine and anthropogenic influences. Moreovel; the distribution of nutrients, organic and heavy metals is highly related to the landuse and development history of local districts. Finally, the rivers in the fringes of New Towns demonstrate multiple types of river pollution.