Optimal design of groundwater pollution management systems for a decanting contaminated site: A simulation-optimization approach

Abstract

Major challenges facing groundwater management arise mostly from approaches aiming at ensuring water resources sustainability, water quality protection and aquifer restoration. In many old polluted sites, the most challenging practice is the management of decanting contaminated water from flooded shaft that decants into the natural environment. Pump and treat (PAT) is one of the commonly used techniques in management of groundwater pollution problem. Simulation-Optimization (S–O) models have proven to be very useful in appropriate design of an effective PAT groundwater remediation system. In this study, simulation models MODFLOW and MT3DMS for groundwater flow and transport respectively, and a solving constraint integer program (SCIP) optimization algorithm were employed in designing of an effective system for cleaning and controlling pollution coming from hypothetical decanting site. Simulation models were used to predict the spatial and temporal variation of contaminant plumes and aquifer response to remediation measures. SCIP optimization algorithm, on the other hand, was applied to determine the optimal solution of the formulated objective function while attaining a set of water quality and head constraints as well as operation bounds. Results from this study indicated that the optimal pumping rates for the proposed pumping wells met water quality and water head requirements. Further, the obtained results demonstrate the application of SCIP in conjunction with simulation models as a capable tool in designing an effective groundwater management system and control decanting problem. The performance of SCIP is also compared with genetic algorithm (GA) and the SCIP is found to have a better convergence rate than the GA.