Assessment of Seasonal and Spatial Variations of Coastal Water Quality Using Multivariate Statistical Techniques

Abstract

This study investigates the seasonal and spatial trends in Kuwait’s coastal water’s physical, chemical, and biological parameters by applying multivariate statistical techniques, including cluster analysis (CA), principal component/factor analysis (PCA/FA), and the Pearson correlation (PC) method to the average daily reading of water quality parameters from fifteen stations over one year. The investigated parameters are pH, turbidity, chlorophyll-a, conductivity, dissolved oxygen (DO), phycoerythrin, salinity, and temperature. The results show that the coastal water of Kuwait is subjected to high environmental pressure due to natural and human interferences. During 2017, the DO levels were below the threshold limit, and at the same time, the water temperature and salinity were very high, causing a series of fish death events. CA resulted in three different regions based on the turbidity, including high, moderate, and low regions, and three seasons (winter, summer, and autumn). Spring is very short and overlaps with winter and summer. PCA/FA applied on the datasets assisted in extracting and identifying parameters responsible for the variations in the seasons and regions obtained from CA. Additionally, Pearson’s correlation resulted in a strong positive relation between chlorophyll and phycoerythrin in 7 out of the 15 stations. However, at high turbidity regions (stations 1 and 2), chlorophyll concentration was low. Additionally, the negative correlation between DO and temperature was observed at stations with rare human activities.