Evaluation of Groundwater Quality Monitoring Network in Khanmirza Plain Using the Discrete Entropy Theory

Quality of recorded information in a monitoring network plays an important role in sustainable design and operation of water resources systems. For an optimal monitoring network design, they should be evaluated periodically based on the information needs and future water resources development plans. In this study, regional values of groundwater quality monitoring network stations in Khanmirza plain were investigated using the discrete entropy theory. Discrete entropy can overcome the limitations regarding data normality assumption in past research. Several entropy parameters such as marginal entropy, joint entropy, information send and received by each station and transinformation index between stations, were calculated to identify essential stations and poor regions based on the amount of information exchanges. Sensitive analysis to number of discrete intervals shows that the entropy indices are sensitive to change of interval length, but the ranks of stations appear to be less sensitive. Validation result confirms that the entropy has a good performance in quantifying the regional values of sampling wells in a monitoring network and comparison with the correlation coefficient of information between stations shows that this coefficient can be used for determining the essential and unnecessary stations in the region.