Optimal Identification of Groundwater Pollution Sources

An inverse problem model of groundwater pollution source identification has been presented in the present study. The model is based upon linked simulation optimization approach and invokes a finite difference based simulator and an Exhaustive Search based optimizer. The model estimated pollution source characteristics are those set of source characteristics which provide the closest match between observed and the simulated concentration field. The pollution source characteristics include number of sources, their locations and release flux rates. Finding the optimum number of sources and their location is an advantage of this model, since most of the source identification problem involves a priori known potential location. The model has been illustrated for a hypothetical study area, with known flow/transport boundary conditions and flow/transport parameters. Two dimensional steady state flow and transient transport has been considered. Identified results indicated that the proposed methodology can be used to solve the inverse problem of groundwater pollution source identification.