Multivariate Statistical Analyses of Various Physico-Chemical Parameters and Selected Metals in Soil in Vicinity of Sugarcane Based Industrial Unit

Abstract

Multivariate analysis is a powerful tool to unravel the hidden aspects of a particular data set. The present investigation deals with the use of multivariate analysis (i.e. principal component analysis and cluster analysis) for exploring the sources of various studied parameters in the soil samples and also determines the pollution status of the soil in vicinity of a sugarcane based industrial unit located in Mandi Bahauddin, Pakistan. The industrial unit comprised of sugar mills and a few chemical industries producing methanol, ethanol etc. from the bye-products of sugar mills. A total of 36 soil samples were collected from the peripheral distance of 50, 100 and 150m from the industrial unit and analyzed for various physico-chemical parameters like pH, electrical conductivity, alkalinity, chloride, sulfide and sulfate. The selected metals (i.e. K, Mg, Co, Ni, Pb) were analyzed by flame atomic absorption spectrophotometer under optimum analytical conditions. The data thus obtained was subjected to univariate and multivariate statistical analyses. The results evidenced the contamination of the soil in vicinity of the industrial unit upto a critical distance of 150 m from effluent discharge point by heavy loads of chloride, sulphide, sulphate and almost all the selected metals. The correlation coefficient matrix revealed a number of strong positive relations among various metal pairs. Multivariate analysis evidenced the sources of metals to be present in various processes of sugar and other industries.