Assessment Of Water Quality Using Multi-Variate Analysis,Water Quality Index And Geo-Statistical Analysis

The water quality is mostly characterized by many variables (parameters) which represent a water composition in specific localities and time. Real hydrological data are mostly noisy, it means that they are not normally distributed, often co-linear or auto correlated, containing outliers or errors etc. These data sets create a n-dimensional space from which information about the water composition has to be mined. For this purpose, multivariate methods such as the Principal Component Analysis(PCA), the Factor Analysis, and the Discriminate Analysis, are used. An alternative is not to use all the quality parameters themselves, but instead of them the indexes that synthesize the typical similarities and differences between the major ions in the surface water environment. These new variables (indexes) can be easily identified by means of a factorial method. Then trends can be analyzed by a robust technique. In the present investigation, an attempt has been made to study Multi-variate statistical analysis of groundwater collected from different sampling points during different seasons. This project objective is to assess existing water quality condition and future water quality conditions for selected parameters. The basic aim is to use PCA for reducing the number of parameters without losing much information which was determined during the three season and to recognize basic feature of water quality. water quality index is one of the most effective tool to communicate the information on the quality of any water body. Geo-statistical technique, namely Ordinary Kriging is been used to enhance the water quality data prediction. This is widely and the most prominent spatial statistical analysis method in Estimation