Based on the fact that there is spatial variation of hydraulic conductivity, a new probabilistic Pareto genetic algorithm (PPGA) is developed to solve multi objective optimal design of groundwater contaminant monitoring network under the spatial variation of hydraulic conductivity. The PPGA is developed by adding the probabilistic Pareto domination ranking and probabilistic niche technique to the classic epsilon dominance non dominated sorted genetic algorithm II (ε NSGAII) to search for Pareto optimal solutions of multi objective optimization problems under uncertainty. The Pareto optimal solutions are then compared with the MC analysis results to demonstrate the effectiveness and reliability of the PPGA. Comprehensive analysis demonstrates that the proposed PPGA can find Pareto optimal solutions with low variability and high reliability and can provide a range of reliable monitoring programs for decision makers under the spatial variation of hydraulic conductivity.