Inexact optimization modeling for water quality management in Xingshan County of the Xiangxi River Basin

The Three Gorges Reservoir (TGR), the largest reservoir in China, has confronted the local development with many environmental challenges. Xingshan County, located in vicinity of the TGR, is one of the areas that have been significantly influenced. It lies within the middle reaches of the Xiangxi River, with a total area of 2,327 km2. In recent years, changes to the County including rapid economic development and interventions of the TGR have brought about many pressing water quality problems. A sustainable development strategy is urgently needed to mitigate water deterioration and sustain economic growth. However, water quality systems are often associated with various uncertainties, which would strongly affect the related decision making processes. Therefore, this research aims to provide scientific support for achieving sustainable development strategy for the County under uncertainties, with a focus on integrated water quality management.;The proposed models can effectively communicate uncertainties expressed as intervals, probabilistic and possibilistic distributions into the optimization processes, and generate mitigation schemes with varied levels of system-failure risk. Results indicate that no pollution control would be required when the incoming water quality is Grade I. However, when the incoming water quality degrades to Grade II or III, certain mitigation should be conducted. This study is a first attempt to apply optimization approaches into water quality management at the watershed level in the TGR. It can provide scientific bases for regional sustainable development and planning.;An inexact two-stage stochastic programming water quality management (ITSP-WQM) model is established for the County. Various environmental, resource, and economic factors are integrated within the optimization framework. Interval solutions are obtained, and four scenarios are analyzed to reflect tradeoffs between the predefined economic targets and the associated environmental penalties. Moreover, in order to further tackle the fuzzy information, an inexact two-stage stochastic credibility constrained programming water quality management (ITSCCP-WQM) model is developed. The proposed model extends upon the ITSP-WQM model by introducing credibility constrained programming, and thus, fuzziness of the pollution load capacities can be reflected. Interval solutions that contain a spectrum of potential pollution abatement options are obtained. Sensitivity analysis regarding the total phosphorus control is conducted.