A multi-objective spatial optimization of wetland for Sponge City in the plain, China

Abstract

Climate change exacerbates various environmental problems in cities, such as flooding, droughts, and water pollution. As a result of these urban environmental problems, China vigorously promotes the construction of Sponge City. Wetlands are an effective stormwater management facility of Sponge City which can regulate urban stormwater runoff and remove non-point source pollution from stormwater runoff. Wetlands, as sensitive ecosystems in cities, require additional consideration and rational planning in the construction of Sponge City. Wetlands constructed in suitable locations can improve constructed wetland potential, improve wetland benefits, and reduce cost. However, the multi-objective spatial optimization of wetlands in Sponge City is insufficient. To fulfill wetlands planning and construction for Sponge City, a multi-objective optimization method that maximizes constructed wetland potential and pollutant removal and minimizes input costs is proposed by genetic algorithm (NSGA-II, Non-dominated Sorting Genetic Algorithm-II). The wetland space was optimized for the study area, and the Pareto optimal solution of the wetland layout was obtained. The values of the three objectives of wetlands spatial optimization schemes (Pareto optimal solutions) identified is compared with the values of the three objectives of the planned wetlands, which verifies the effectiveness of the wetland spatial optimization method. On the basis of the Pareto optimal solutions, we propose four schemes with the highest economic benefit scheme for constructed wetland potential (constructed wetland potential/input costs), highest economic benefit scheme for pollutant removal efficiency (pollutant removal/input costs), the best compromise scheme between three objectives, and highest constructed wetland potential scheme. Compared with the planned wetlands scheme, these four schemes can increase the constructed wetland potential by 2.2% ∼ 25.1%; increase the TN mass removal by −8.8% ∼28.0%; reduce the input costs by 8.9% ∼ 37.1%.