A NEW SIMULATION-OPTIMIZATION APPROACH FOR SIMULTANEOUSLY IDENTIFYING THE SPATIAL DISTRIBUTION AND SOURCE FLUXES OF THE AREAL GROUNDWATER POLLUTION SOURCES

This study proposes a new simulation-optimization approach for simultaneously identifying the spatial distribution and source fluxes of the areal pollution sources in groundwater systems. In the proposed approach, groundwater flow and pollution transport processes are simulated via MODFLOW and MT3DMS models in the simulation part. These models are then integrated to an optimization model where a binary genetic algorithm (GA) is used. In the proposed GA based optimization model, finite difference grid blocks of the given aquifer domain are considered to be the potential areal pollution source locations. The main objective of the GA is to evolve the source fluxes and spatial distributions of source locations through genetic operators by minimizing the error value calculated between the measured and simulated pollution concentrations at given monitoring locations and times. The performance of the proposed approach is evaluated on a hypothetical aquifer model for 4 different pollution source distributions. Identified results indicated that the proposed simulation-optimization approach may be used as an effective way to solve the areal pollution source identification problems.