Search

A simulation-based interval stochastic bi-level multi-objective programming (SISBLMOP) model was proposed in this research, through integrating the global nutrient export from watersheds model, interval parameter programming and stochastic chance-constrained programming into a general bi-level multi-objective programming framework. The SISBLMOP model can handle multiple uncertainties expressed as discrete intervals and probability density functions in both the simulation and optimization processes. System complexities, including the hierarchy structure of upper- and lower-level decision makers, can also be addressed in the model. The proposed model is applied to a real-world case study of the Xinfengjiang Reservoir Watershed in South China to identify the satisfactory implementation levels of multiple best management practices (BMPs). The model results show that multiple BMP schemes for water quality management can be obtained under different upper- and lower-level decision-making and risk-violation scenarios, reflecting the cooperation and gaming results of the two-level decision makers. Consequently, the corresponding BMP implementation costs are acceptable to both the upper- and lower-level decision makers. The model is widely applicable and can be effectively used for water quality management under multiple uncertainties and complexities. • A simulation-based interval stochastic bi-level programming model was developed. • System hierarchical structure and multi-objective characteristics were considered. • The stochastic characteristics of nutrient export can be reflected. • The model can reflect uncertainties in both simulation and optimization processes. • The model can support watershed water quality management. [ABSTRACT FROM AUTHOR]

The urbanization process has seen an accelerated increase in recent decades, leading to urban runoff pollution becoming more prominent. However, uncertainty of the pollution output and complexity of management systems have made controlling urban runoff pollution challenging. Therefore, it is necessary to propose advanced modeling methods for these challenges. This research presents an integrated urban runoff pollution management (IURPM) model for optimal configuration of low impact development (LID) practices under multiple uncertainties. The IURPM model combines the hybrid land-use prediction and improved pollution estimation models with interval parameter, stochastic parameter, and multi-objective programming. The proposed IURPM model can not only predict the output characteristics, but also provide optimal configuration schemes for the LID practices in the management of urban runoff pollution under multiple scenarios. In addition, uncertainties expressed as discrete intervals and probability density function in the management systems can be effectively addressed. A case study of the IURPM model was conducted in Dongguan City, South China. Results show that considerable amounts of urban runoff pollutants would export from Dongguan City by 2025. The export loads and pollution output flux per unit area would have significant spatial heterogeneity. The results further indicate that population size, gross domestic product, and regional area size are expected to play important roles in the pollution export, while impervious surface coverage and population density would likely have great influences on the output flux of urban runoff pollution. Based on the model findings, multiple LID practices should be adopted in Dongguan City to reduce the urban runoff pollution loads. Using the IURPM model, multiple LID implementation schemes can be obtained under different pollution reduction scenarios and significance levels, that can provide decision-making support for urban water environmental management, considering variations in the policymaker's decision-making preferences. This study demonstrates that the IURPM model can be applied to the optimal configuration of LID practices for the management of urban runoff pollution under uncertainty. • An integrated urban runoff pollution management model was proposed. • The model can predict the loads of urban runoff pollution. • The model can provide optimal configuration schemes of low impact development practices. • Interval and stochastic uncertainties in urban runoff pollution systems can be handled. [ABSTRACT FROM AUTHOR]