Sequential Optimal Monitoring Network Design using Iterative Kriging for Identification of Unknown Groundwater Pollution Sources Location

Identification of unknown groundwater pollution source characteristics, in terms of location, magnitude and activity duration is important for designing an effective pollution remediation strategy. Precise source characterization also becomes very important to ascertain liability, and to recover the cost of remediation from parties responsible for the groundwater pollution. Due to the uncertainties in accurately predicting the aquifer response to source flux injection, generally encountered sparsity of concentration observation data in the field, and the non uniqueness in the aquifer response to the subjected hydraulic and chemical stresses, groundwater pollution source characterization remains a challenging task. A scientifically designed pollutant concentration monitoring network becomes imperative for accurate pollutant source characterization. The efficiency of the unknown source locations identification process is largely determined by locations of monitoring wells where the pollutant concentration is observed. The proposed method combines spatial interpolation of concentration measurements and Simulated Annealing as optimization algorithm to find the optimum locations for monitoring wells. Initially, the observed concentration data at few sparsely and arbitrarily distributed wells are used to interpolate the concentration data for the aquifer study area. The concentration information is passed to the optimization algorithm (decision model) as concentration gradient which in turn finds the optimum locations for implementing the next sequence of monitoring wells. Concentration measurement data from these designed monitoring wells and already implemented monitoring network are iteratively used as feedback information for potential groundwater pollution source locations identification. The potential applicability of the developed methodology is demonstrated for an illustrative study area.