Use of statistical and geostastatical methods at selection of water quality sampling points in groundwater

Monitoring studies are important in terms of protecting existing resources, securing human health and ensuring environmental sustainability. Within the scope of this study, alternative water quality monitoring networks were designed by using statistical and geostatistical methods in the selection of groundwater water quality sampling points located on Akdeniz University campus. The field study was conducted on June 23, 2022 at 21 active sampling points located on the Akdeniz University campus. Dissolved oxygen, oxygen saturation, temperature, pH, conductivity, turbidity and nitrate analyzes were performed on the samples collected from the wells. In the study, SPSS (Statistical Package for the Social Sciences) program, which is frequently preferred in statistical analysis, was used for the hierarchical clustering method. ArcMap software, a geographic information systems program, was used to apply the Ordinary Kriging method. Hypothetical monitoring networks were created by selecting sampling points from the clusters formed by hierarchical clustering analysis. There are 38 monitoring networks in the study, including the main monitoring network and 37 hypothetical monitoring networks. Point reduction was made by evaluating variograms, kriging prediction maps and kriging error maps in the monitoring networks, and an alternative network with similar characteristics to the main network was selected. The results demonstrated that 10 sampling points are necessary in the groundwater quality representative monitoring network of Akdeniz University campus to maintain the spatial correlation structure of the main monitoring network. It is concluded that the IA28 monitoring network chosen as the alternative network with the Nugget/Sill ratio is less than 0.25 has a high spatial correlation and it is the monitoring network that best reflects the main network.