Abstract Unplanned and excessive utilization of groundwater increases the risk of seawater intrusion in coastal areas. Therefore, water quality management and monitoring in these areas are significantly important. For designing a monitoring network, the least number of monitoring wells with the optimal spatial distribution should select due to economic considerations. In this study, the optimal monitoring network with the minimum number of wells was selected in the coastal aquifer of Talesh County by considering the aquifer vulnerability maps and assessing the accuracy of the designed monitoring network. Accordingly, the aquifer vulnerability map was prepared using the modified GALDIT index, and then a genetic algorithm was used for optimal search in the monitoring network. The optimization model simultaneously analyzed three objectives: (1) to maximize the correlation between the vulnerability index and the EC value, (2) to minimize the number of monitoring wells, and (3) to maximize the Nash-Sutcliff coefficient that indicates the goodness of fit between the distribution of calculated EC in the existing monitoring network and the new network. The three objectives were integrated into one objective function by applying the weight coefficient w for economic reasons, and then the various weights were assessed. The results showed that the optimal solution selection significantly depended on determining the weight coefficient, and the best weight coefficient was selected by taking the most balanced solution into account according to the vulnerability index and the accuracy of the monitoring network. Satisfied predictions were achieved in both the optimization and the validation steps. Moreover, due to the qualitative and quantitative changes in groundwater in the long term, it should be assessed and redesigned the groundwater quality monitoring network periodically for the monitoring network to be effective in planning and applying methods for improving groundwater quality.
展开▼
