摘要
Water resources availability has a significant impact on agricultural land-use planning, especially in a water shortage area such as North China. The random nature of available water resources and other uncertainties in an agricultural system present risk for land-use planning and may lead to undesirable decisions or potential economic loss. In this study, an inexact risk management model （IRM） was developed for supporting agricultural land-use planning and risk analysis under water shortage. The IRM model was formulated through incorporating a conditional value-at-risk （CVaR） constraint into an inexact two-stage stochastic programming （ITSP） framework, and could be used to control uncertainties expressed as not only probability distributions but also as discrete intervals. The measure of risk about the second-stage penalty cost was incorporated into the model so that the trade-off between system benefit and extreme expected loss could be analyzed. The developed model was applied to a case study in the Zhangweinan River Basin, a typical agricultural region facing serious water shortage in North China. Solutions of the IRM model showed that the obtained first-stage land-use target values could be used to reflect decision-makers＇ opinions on the long-term devel- opment plan. The confidence level a and maximum acceptable risk loss fl could be used to reflect decision- makers＇ preference towards system benefit and risk control. The results indicated that the IRM model was useful for reflecting the decision-makers＇ attitudes toward risk aversion and could help seek cost-effective agricul- tural land-use planning strategies under complex uncer- tainties.
Water resources availability has a significant impact on agricultural land-use planning, especially in a water shortage area such as North China. The random nature of available water resources and other uncertainties in an agricultural system present risk for land-use planning and may lead to undesirable decisions or potential economic loss. In this study, an inexact risk management model （IRM） was developed for supporting agricultural land-use planning and risk analysis under water shortage. The IRM model was formulated through incorporating a conditional value-at-risk （CVaR） constraint into an inexact two-stage stochastic programming （ITSP） framework, and could be used to control uncertainties expressed as not only probability distributions but also as discrete intervals. The measure of risk about the second-stage penalty cost was incorporated into the model so that the trade-off between system benefit and extreme expected loss could be analyzed. The developed model was applied to a case study in the Zhangweinan River Basin, a typical agricultural region facing serious water shortage in North China. Solutions of the IRM model showed that the obtained first-stage land-use target values could be used to reflect decision-makers＇ opinions on the long-term devel- opment plan. The confidence level a and maximum acceptable risk loss fl could be used to reflect decision- makers＇ preference towards system benefit and risk control. The results indicated that the IRM model was useful for reflecting the decision-makers＇ attitudes toward risk aversion and could help seek cost-effective agricul- tural land-use planning strategies under complex uncer- tainties.
