Evaluating spatial and temporal variation in Tuzaklı Pond water using multivariate statistical analysis

Abstract

This study used multivariate statistical techniques to demonstrate the spatial and temporal changes in water quality, main pollutant sources and water quality classes in Tuzaklı Pond. The water quality datasets are obtained on a monthly basis (November 2014–October 2015) using the results of 28 parameters that are obtained from three stations in the pond. Datasets are spatially and temporally assessed using statistical techniques, including one-way analysis of variance (ANOVA), Pearson’s correlation, hierarchical agglomerative cluster analysis (HCA) and principal component analysis (PCA). PCA indicates the four main components responsible for the data structure, accounting for 88.31% of the total variance of the dataset. These main components are physical parameters, soluble salts (natural), ammonium and phosphorus (agricultural activity), which are nutrient elements. Furthermore, it can be temporally concluded using HCA that the summer and autumn seasons exhibit more similar characteristics as compared to those exhibited by the remaining seasons. According to the water quality and class criteria of Turkey Surface Water Management Regulation and the World Health Organisation (WHO), while this pond generally represents Class I, we observed PO43−, SO32−, NO2− and NO3− (Class II), which resulted in slightly contaminated water.