Identification of Contaminant Sources in Aquifers under Uncertainty

The identification of contaminant sources in aquifers is an important issue for the protection of groundwater resources. Recently, various models, including analytical and numerical solutions, have been proposed to solve this problem. In most of these applications the contaminant source identification problem was analyzed using deterministic models. For example, in these applications it was assumed that the aquifer is homogeneous and it has a well-defined hydraulic conductivity distribution. In this study, the parameters such as the hydraulic conductivity and dispersion coefficients are considered to be uncertain parameters. In this study we represent this uncertainty using fuzzy sets. The fuzzy membership functions used in this representation may be based on field data or expert knowledge. Using this approach, the contaminant source identification problem is formulated as an optimization model with fuzzy parameters. The solution is obtained by coupling the genetic algorithm (GA) and vertex method as it is applied to the solution of the optimization model. A hypothetical aquifer is used to demonstrate the effectiveness of the model and the algorithm. The results show that the proposed model can reliably identify the locations of contaminant sources and their release histories for uncertain parameters and the proposed algorithm is shown to be efficient.