Akram Seifi; Hossien Riahi
Abstract
Introduction: Industrial and agricultural activities resulting in the production of toxic heavy metals may endanger water quality, public health, and the environment. Therefore, the determination of areas that are affected by heavy metals and spatial uncertainty of pollution risks are considered as an ...
Read More
Introduction: Industrial and agricultural activities resulting in the production of toxic heavy metals may endanger water quality, public health, and the environment. Therefore, the determination of areas that are affected by heavy metals and spatial uncertainty of pollution risks are considered as an important and sensitive issue, which are less studied. The main aim of this study was to combine Bayesian network analysis with Sequential Gaussian Simulations (SGS) to evaluate the pollution risk of heavy metal and toxic elements in the surface water of Sarcheshmeh copper mine. Material and methods: In this study, a dataset of 924 water samples from 82 locations from three different zones including the surface water of Shour River, tailing dam, and also the main mining site of Sharcheshmeh copper complex and nine heavy metals were used. The information was classified into two risk classes of low and high according to the standard of the Department of Environment of Iran. A Bayesian analysis and learning algorithm were applied to investigate the characterization of heavy metal correlations and Bayesian weights extraction. Based on the obtained Bayesian network structure, important metals were chosen as key pollution parameters. For these metals, the conditional probability was dedicated to every observed point and then the Bayesian Risk Index (BRI) was calculated as a linear rating of the weighted risk classes. Finally, the geostatistical modeling and SGS were applied for generating pollution risk and standard deviation maps of BRI were used as an uncertainty measure of SGS based on BRI elements. Results and discussion: Based on the results of Bayesian analysis, three elements of Zn, Mo, and Fe were identified as the most important parameters of pollution risk in the studied zones, which were derived by the MWST Bayesian network. The highest risk existed in the main mining zone and sedimentation dam. The results of BRIzn, BRIMo, and BRIFe declared that areas in north and south of zone 1 and all of zone 2 had high pollution risk, which requires appropriate treatment operations. The results also showed that the high-risk cluster was mainly located within the main mining and tailing dam zones. Also, 19% and 22% of zones’ area was classified as high and low risk of water pollution, respectively. Zoning maps of risk and heavy metals showed that there are high standard deviation and great variation in copper complex and distilling dam. The results of the uncertainty risk assessment showed high concentrations of heavy metals in the surface water arose from the transportation of heavy metal from copper mine to distilling dam, which requires treatment operation on the output water of the factory. Conclusion: Based on the results, the pollution of heavy metal and toxic elements in water resources near Sarcheshmeh copper mine and downstream water resources was high and this will increase the pollution risk of Rafsanjan aquifer. These indicate the inadequate treatment of heavy metals in Sarcheshmeh copper mine water.
