Source identification and evaluation of surface water quality using factor and discriminant analysis

Abstract

This study utilizes factor and discriminant analysis to identify the parameter source and evaluate the quality of water from ex-mining ponds and lakes in Selangor. Factor analysis (FA) which explain 83.77% of the surface water quality variation shows that As and Cd that mostly originated from mining activity, and pH are the parameters responsible for the major variation in the surface water quality and were strongly associated with varimax factor 1 (VF1), while Pb, Mn and DO were associated with varimax factor 2 (VF2). The Discriminant analysis (DA) reveals that As, Cd, Mn, Fe and pH are the parameters that significantly differentiate ex-mining ponds from the lake (p < 0.05 and F - ratio >> 1), and supported by the correlation study. The elevated metal concentrations in ex-mining ponds compared to lakes were plausibly associated with the past mining operation, thereby indicating heavy metal dominance in ex-mining ponds. DO and BOD are associated with the anthropogenic input from residential sources. Findings of this study therefore show the need and usefulness of multivariate statistical analysis to get information on the quality status of surface water.

Keywords: Discriminant analysis, Ex-mining pond, Factor analysis, Lake, Water quality.