Determining the origin of arsenic anomalies in groundwater using multivariate statistical methods (case study: Miandoab plain aquifer, NW of Iran)

Geological and hydrogeological processes are complex and in the many measurements, this complexity with nonuniform distribution becomes more visible. Usually, temporal and spatial measurements of physical or chemical processes cannot detect the main dominated processes. The data reduction of multivariate analysis, try's to classify, interpretation and detecting the main govern processes, which in this present study were used to determining the origin of hydrochemical anomalies. To assess the major and minor elements of Miandoab plain, 54 groundwater samples were collected. In addition, 36 groundwater samples were collected and analyzed for assessing the trace elements in the study area. Factor analysis, Pearson correlation, and Hierarchical clustering methods were used to analyzing the hydrochemical results. According to the results of the Multivariate statistical methods and due to the high correlation of arsenic with iron and silica, the origin of arsenic is attributed to the dissolution of rocks and sediments in the region. The presence of Sahand volcanic mountains in the area and the spread of igneous rocks are contributing to the arsenic anomaly in the area. Based on the cluster analysis, samples S3, S4, S5, S14, and S18, which have a more anomaly of trace element, were placed in one subcluster and indicate the geogenic origin of As, Fe, and SiO2. Also, samples S7 and S20 with more anomaly of arsenic can represent the anthropogenic origin of arsenic. The use of herbicides and pesticides in the study area has been identified as a major cause for anomalous concentrations of arsenic in these samples.