Use Of Cluster Analysis Technique For Water Quality Assessment In Mahitsy Commune, Central Highland Of Madagascar

Research Article
Mamiseheno Rasolofonirina., Voahirana Ramaroson and Solofonirina Dieu Donné Ravelomanantsoa
DOI: 
http://dx.doi.org/10.24327/ijrsr.2018.0901.1416
Subject: 
science
KeyWords: 
Water quality, physico-chemical, parameter, cluster analysis, trilinear diagram.
Abstract: 

This study assessed the characteristics of well waters sampled in Mahitsy Commune using Piper trilinear diagram and cluster analysis to determine the similarities or dissimilarities among the sampling sites. The linear diagram showed that nine of fifteen water samples were of saline water type with sodium and chloride as dominant cation and anion, respectively. Five water samples were classified as bicarbonate type waters. Depending on the sampling points, calcium, magnesium or sodium was found to be the dominant cation for the bicarbonate water samples. Data clustering analysis using the hierarchical agglomerative method grouped the fifteen sampling points into three main clusters (A, B and C) based on the similarity of the water quality characteristics. The cluster groups A and B comprised nine water samples, which were of saline type and contained high concentrations of nitrate (more than 260 mg/l as NO3 - ). The cluster group C was made up of six water samples, which were all characterized by low TDS, except for one sample with a high concentration of nitrate (117 mg/l as NO3 - ).