Objective  The stormwater problem has become a key restrictive factor for the development of shallow mountain areas and the low impact development (LID) of greenspace is an important means to solve such problem in shallow mountain areas. Forming an optimization method for distribution of LID controls for multi-objectives, such as runoff control and cost, can provide important technical support for the efficient solution of stormwater problems in shallow mountain areas, and contribute to the future high-quality development of these areas.

  Method  Based on the characteristics of greenspace planning and design and runoff in shallow mountain area, the study formed a platform for optimal distribution of LID controls by D8 and NSGA-Ⅱ coupled algorithm, which realized spatial quantitative optimization of the type and scale of LID controls based on collaborative optimization of runoff control and cost. In addition, Westmount Country Park in Shijiazhuang City was taken as the experimental object to verify the feasibility of the method.

  Result  24, 30 and 30 optimal solution sets for the optimal distribution of LID controls of the study area were obtained in the simulated 2-h rainfall event under return periods of 5-year, 10-year and 20-year; the ‘ideal investment upper limit point’ of each simulated rainfall event was 75.141 million CNY, 66.344 million CNY and 60.651 million CNY, respectively; visual results of the distribution of most efficient cost based on ArcGIS showed that raingarden, permeable pavement and water were scattered, and vegetative swale was scattered in small-scale linear distribution.

  Conclusion  D8 and NSGA-Ⅱ coupled algorithm can well match the LID of greenspace in shallow mountain area, and simplify the cumbersome design process of LID in traditional greenspace; there is a diminishing marginal benefit between peak flow of runoff and LID controls cost, which accelerates with the increase of rainfall return period; the permeable pavement and raingarden have more cost performance than other LID controls; the visual simulation results of the experimental object basically accord with the principle and pattern in real design, which verifies the feasibility and rationality of the method; in order to improve the rationality and guidance of simulated results of distribution, the further research should focus on the collaborative method for optimal distribution of LID controls and greenspace planning and design.