Use of Principle Component Analysis (PCA) to Evaluate the Groundwater Resources of Havsa District (Edirne, TURKEY)

Abstract

Havsa District is located in the Edirne Province in Thrace Region of Turkey and provides drinking water to about 30.000
people. This study was carried out to evaluate the groundwater quality of Havsa District by using some multivariate statistical
techniques. For this purpose, groundwater samples were collected from 15 stations from the villages of Havsa District in
winter season of 2016. Some water quality parameters including dissolved oxygen, % oxygen saturation, total dissolved
solids (TDS), salinity, electrical conductivity (EC), pH, sulphate (SO4), nitrate (NO3), phosphate (PO4) and chlorine (CI) in
groundwater of region were determined and Factor Analysis (FA), Cluster Analysis (CA) and Pearson Correlation Index
(PCI) were applied to detected data in order to evaluate the groundwater quality. According to the results of PCI, significance
positive and negative relations were determined among the investigated parameters at the 0.05 and 0.01 levels. According to
results of FA, 2 factors named as "Agricultural Factor" and "Oxygenation Factor" explained 87.16% of the total variance.
According to CA, 2 statistically significant clusters were formed, which were corresponded to polluted and unpolluted stations.