A multivariate statistical approach for monitoring of ground water quality: a case study of Beri block, Haryana, India

摘要

The present research deals with assessment of groundwater quality of Beri block, Jhajjar district, Haryana, India and its nearby villages. Multivariate statistics is an efficient technique to display relationship between different limiting factors. Around 24 groundwater samples were collected. A total of 16 variables were analysed: pH, potassium, total dissolved solids (TDS), hardness (calcium, magnesium and total), sulphate, sodium, electrical conductivity and phosphate, chloride (Cl-) and heavy metals, namely iron, chromium, lead and zinc. Principal component analysis is one of the commonly used tools in water quality assessment because it effectively reduces number of variables. Multivariate statistical tools "principal component analysis (PCA)" and "cluster analysis" were used to set up relationship among the studied parameters. PCA showed the existence of up to five significant PCs which account for 80.35% of the variance. Few parameters such as pH, sodium, potassium, sulphate, phosphate and zinc were found to be well within limits as approved by WHO and BIS, whereas parameters such as chloride, alkalinity, hardness, total dissolved solids and metals (Pb, Cr and Fe) were found to go beyond the prescribed limits. High levels of hardness, total dissolved solids and chlorides are responsible for saline behaviour of water. The correlation matrices for 16 parameters were executed. EC, TDS, chloride and total hardness were significantly and positively correlated with each other. pH and phosphate (PO42-) were negatively correlated with majority of the physicochemical variables. After studying the physiochemical properties of groundwater samples, it is recommended that water quality parameters should be analysed periodically to conserve the water resources and emphasis should be laid on water quality management practices.