Water Quality Assessment of Muar River Using Environmetric Techniques and Artificial Neural Networks

• Main Author: Putri Shazlia, Rosman
• Format: Conference or Workshop Item
• Language: English
• Published: 2015
• Subjects: T Technology (General)
• Online Access: http://umpir.ump.edu.my/id/eprint/10221/1/Water%20Quality%20Assessment%20of%20Muar%20River%20Using%20Environmetric%20Techniques%20and%20Artificial%20Neural%20Networks.pdf; http://umpir.ump.edu.my/id/eprint/10221/

The pollution discharge has influence the chemical composition of Muar River where studied was carried out using the Environmetric Techniques and the Artificial Neural Networks (ANNs) model. Environmetric method, the hierarchical agglomerative cluster analysis (HACA), the discriminant analysis (DA), the principal component analysis (PCA) and the factor analysis (FA) to study the spatial variations of water quality variables and to determine the origin sources of pollution. ANNs model was used to predict linear relationship between water quality variables, the most significant variables that influence Muar River as well as sources of apportionment pollution. HACA observed three spatial clusters were formed. DA managed to discriminate 16 and 19 water quality variables thru forward and backward stepwise. Eight principal components were found responsible for the data structure and 67.7% of the total variance of the data set in PCA/FA analysis. ANNs analysis, strong relationship correlation was observed between salinity, conductivity, DS, TS, Cl, Ca, K, Mg and Na (r = 0.954 to 0.997), moderate relationship observed between COD and E.coli (r = 0.449) and Cd and Pb (r = 0.492) and others variables have no significant correlation. pH was the most significant variables (51.6%) and Fe was less significant variables (-0.52%). The major sources of pollution of the river were due to natural degradation / natural process that affecting the pH value of the river. Other pollution contribution was from anthropogenic sources such as agricultural runoff, industrial discharge, domestic waste, natural erosion, livestock farming and present of nitrogenous species. The ANNs showed better prediction in identified most significant variable compare to Environmetric techniques. ANNs is an effective tool in decision making and problem solving for local/global environmental issues.