In recent decades, population growth, agricultural development and climate change have caused environmental problems in the world, especially in the arid and semi-arid regions such as Iran. Urmia Lake, one of the unique biosphere reserves of the world, has suffered from these problems in the last 20 years. Although many studies conducted to the negative impacts of human-induced activities, especially agricultural expansion on the lake, few studies focused on providing restoration plans for that. This study uses a system dynamics method for modelling the restorations scenarios of Urmia Lake based on the prey-predator approach. This method is helpful to simplify the complicated feedbacks and interactions among human-environment. The Monte Carlo simulation is used to explore the effect of uncertain parameters on the lake level. Based on the results of this study, the agricultural water consumption (as predator) has the most significant effect on reducing the lake volume (prey). The results indicated that decreasing the area of irrigated lands by 40%, improving irrigation efficiency, reducing the lake area itself, and inter-basin water transfer will lead to increase in the lake water level by 2.54 m, 2.04 m, 1.19 m and 1.19 m, respectively. By applying all the suggested scenarios, the lake water level will reach to its ecological level (1274 MSL) in 2023. The recent positive restoration signals of the lake have coincidence with the findings of current research and bring more hope to the future of the lake.