Management of nutrient export from diffuse sources in watersheds for environmental protection under uncertainty

Abstract

Nutrient export from diffuse sources poses a significant threat to watersheds, and methods to support the effective management of these watersheds is essential. However, real-world nutrient flows and watershed management systems are highly complex, with a high degree of uncertainty in their descriptive information. Thus, an effective optimization method for watershed management must be developed to deal with this uncertainty, which is crucial for formulating and implementing appropriate management practices. This research presents an inexact simulation-based left-hand-side chance-constrained mixed-integer programming (ISLCCMIP) model to determine nutrient export characteristics and optimal management strategies. By introducing interval and stochastic parameters into the simulation process, the uncertain characteristics of nutrient export loads can be considered. Uncertainties and complexities in management processes can also be handled through incorporating interval parameter programming and mixed-integer programming within a left-hand-side chance-constrained programming framework. The proposed ISLCCMIP model can correlate the randomness in the simulation process and the optimization results. The East River basin in South China was selected as the case study area to apply the proposed model. The results indicated that inorganic nitrogen (N) and phosphorus (P) were the main forms for the nutrient export from diffuse sources in this basin. Five of the nine subbasins were identified as critical source areas for nutrient export. Planting areas of different crops, application amounts of chemical fertilizers, and quantities of livestock types can be optimized to achieve the maximum economic benefit under limited N and P discharge permits. Particularly, planting areas of vegetable and soybean (Glycine max L.) would be first decreased, while rice (Oryza sativa L.) and tubers would be retained, as the pollution emission standards become stricter. Decrease in the planting areas of soybean and vegetables would effectively decrease the violation risk under particular nutrient discharge standards. Concurrently, application amounts of N and P fertilizers and numbers of livestock and poultry should be reduced with stricter pollution discharge standards and increasing significance levels. Furthermore, several strategies correlated to multiple constraint-violation risks, and nutrient reduction scenarios were formulated. These results can provide alternatives for watershed decision-makers to balance increasing system economic benefits and decreasing diffuse source pollution. The proposed ISLCCMIP model can also be applied to other watersheds with similar concerns.