Optimizing best management practices for nutrient pollution control in a lake watershed under uncertainty

• Main Authors: Dai, Chao, Qin, Xiaosheng, Tan, Q., Guo, H. C.
• Other Authors: School of Civil and Environmental Engineering
• Format: Article
• Language: English
• Published: 2018
• Subjects: Fuzzy Credibility Chance-constrained; Programming
• Online Access: https://hdl.handle.net/10356/88732; http://hdl.handle.net/10220/44771
• Institution: Nanyang Technological University

In this research work, soil and water assessment tool (SWAT) and fuzzy credibility chance-constrained programming (FCCP) model were integrated into an optimization framework for supporting nutrient pollution control in a lake watershed. The framework, so-called SWAT-based FCCP (SFCCP) model, was advantageous in simulating non-point sources (NPS) pollution, optimizing best management practices (BMPs), and addressing system uncertainties. SFCCP was solved by genetic algorithm (GA) for searching optimal placement schemes of BMPs at a lower system cost, where the related uncertainties were addressed as fuzzy parameters. The developed SFCCP model was applied to seek optimal types, sizes and locations of BMPs for nutrient pollution control in a lake watershed system, i.e. Lake Dianchi watershed, China. The study results indicated that when the credibility level increased from 0.55 to 0.95, the total system cost would increase from $8.89 to $12.27 × 106; meanwhile, the total nitrogen (TN) load discharged into the lake would decrease from 1.22 to 1.19 × 106 kg/yr and the total phosphorus (TP) load would reduce from 51.37 to 50.11 × 103 kg/yr, respectively. It appeared that a higher credibility level would lead to a stricter control requirement, namely a higher reduction of nutrient load by BMPs and a higher system cost. The proposed model could be used to help generate a series of BMPs placement schemes under various credibility levels; this ensures that the nutrient load brought into the lake and tributaries could drop to an acceptable level, with a proper tradeoff between system cost and risk being considered.