Interrelationship between geochemical elements of sediment and groundwater at Samrak Park Delta of Nakdong River Basin in Korea: multivariate statistical analyses and artificial neural network approaches

This research was conducted at Samrak Park Delta of Nakdong River Basin in Busan Metropolitan City, Korea. The main objective of this study was to evaluate the interrelationship of geochemical elements in sediments and groundwater through multivariate statistical analyses and a multilayer perceptron artificial neural network model. The mean concentrations of chemical elements were Si (46%), Fe (16.9%), Al (15.7%), K (7.5%) and Ca (4.5%) in sediments, and Na (8650 mg/L), Mg (999 mg/L), Ca (432 mg/L), K (293 mg/L) and Cl (17,640 mg/L) in groundwater, respectively. The principal component analysis produced 3 kinds of factors with the variances of 63.37, 27.02 and 9.62%, respectively. It is suggested that the chemical components of sediments and groundwater were mainly originated from source rocks and seawater intrusion, with the minor impacts of irrigation and industry. Cluster analysis also showed that chemical elements were mainly controlled by the natural geogenic sources and seawater intrusion. Multilayer perceptron of artificial neural network (ANN) presented the good interrelationship between sediment and groundwater. The determination coefficients (R ²) between ANN predicted values and observed values in groundwater showed the high values of 0.61–0.97 except Mg, Mn and Sr. It is revealed that the chemical components of sediment and groundwater were derived from local geological origin and from the minor impact of anthropogenic sources. Multivariate analyses and ANN contributed to the identification of the mutual relationship between the geochemical elements of sediment and those of groundwater.