A conditional value-at risk-based two-stage stochastic programming model for grassland ecosystem restoration—A case study for the three-river source

Abstract

Taking three-river source grassland ecosystem as an example, a conditional value-at risk-based two-stage stochastic programming model was proposed for grassland ecosystem restoration. Overload area and restoration area of grassland ecosystem in the three-river source region can be optimized with consideration of environmental requirements, soil erosion and restoration cost. Loss of overload, increased ecological benefit and additional cost of restoration can be evaluated, which will provide theoretical basis for grassland ecosystem restoration and management in three-river source region. Results indicate that overload area and loss of overload will decrease with the stochastic level increase. The total ecological benefit of grass ecosystem will decrease with risk level decrease, indicating that risk attitudes of decision-makers play a decisive role in developing sustainable ecological management schemes. The obtained results can be helpful for decision makers to balance the conflicts between ecological benefit and system risk. Reducing the total amount of livestock should be taken as the main goal of grassland ecosystem restoration and management, while ecological stockbreeding should be developed to promote sustainable development of grassland ecosystem and livestock. Besides, compensation mechanism in related areas should be refined to coordination the relationships between economic development and ecosystem conservation.