Objective The General Office of the Ministry of Housing and Urban-Rural Development clearly proposed that the construction of sponge city should focus on the integration of multiple objectives, emphasize the optimization of life cycle design, and advocate scientific hydrological performance evaluation of completed projects, which requires us to focus on built-up areas, build a more suitable and economical sponge facility system with higher comprehensive performance and establish and improve the evaluation mechanism for construction performance of sponge city, so as to usher sponge city into the construction optimization and scientific evaluation stage. According to the pilot requirements of sponge city, more than 80% of the built-up areas need to reach the sponge city construction target by 2030. In this context, the number of spongy residential areas, which apply the concept of sponge city to build sponge facilities to achieve the target requirements of runoff control, is on the rise. At present, there is a lack of research on long-term hydrologic performance evaluation of spongy residential areas in China. Most researches focus on hydrological performance under short-term rainfall scenarios, or on the elaboration of sponge technology design methods for completed projects. Besides, the verification process of hydrology model in the research on spongy residential areas is relatively lacking. Most researches adopt the outflow monitoring data spanning 1-2 short rainfall periods for model verification, so the verification results may be accidental. It is necessary to adopt the outflow monitoring data spanning multiple long rainfall periods for model verification, so as to improve the reliability of model results. In addition, there is a lack of post-evaluation of the cost consumption of sponge systems in completed projects in China. There are few researches on how to reduce the construction and maintenance cost of sponge system in residential areas. The aim of this research is to explore the design approach for typical sponge systems for residential areas in regions with subtropical monsoon humid climate through the quantitative evaluation of long-term hydrology-cost performance.

Methods Taking the demonstrative spongy residential area in Chongqing as the research object, this research adopts EPA Storm Water Management Model 5.1 to build the current hydrological model. Modeling parameters include hydrological parameters of the underlying surface in the catchment area, parameters of the drainage pipe network, and structural parameters of sponge facilities. The model is accurately verified using monitoring data, and Nash-sutcliffe efficiency coefficient is used to test the matching degree between the monitored runoff value and the simulated model value. The data from rain gauges set around the residential area (stable data from April to October 2018) and the discharge monitoring data corresponding to the rainfall events of the residential area are used to select events with rainfall duration of more than 6 hours for parameter calibration and model verification. After parameter calibration, the monitoring data of typical rainfall events is selected and input into the model to verify the simulation results. The results reveal that the E_NS and R² values are both above 0.75, indicating a high matching degree between the simulation results and the monitoring results, and that the current hydrological model is reliable. In order to explore the optimization model of sponge system design for typical residential green space, the research adopts two sponge system comparison schemes S1 and S2 (the current sponge system scheme is S0), with a view to improving the hydrology-cost performance of the sponge system. Both S1 and S2 are expected to achieve higher hydrological performance and lower cost than S0. On the basis of the current hydrological model already monitored and verified, the sponge system is reset according to S1 and S2 schemes. The hydrological model of the selected schemes is constructed to quantitatively evaluate the long-term hydrological performance, cost input and comprehensive hydrology-cost performance of each scheme.

Results The results show that the total annual runoff control rate of the current scheme is 75.3%, which is slightly lower than the planning control target, and the construction and maintenance cost of the current sponge system is almost the highest. As to scheme S1, the average annual runoff total control rate is 87.2%, the peak reduction efficiency is slightly lower than the current scheme, and the total input cost of the sponge system is the lowest. As to scheme S2, the average annual runoff total control rate is the best, reaching 89.7%, the peak reduction rate of rainstorm (80.3 mm) is 92.8%, and the long-term comprehensive hydrology-cost performance is 3.6 times higher than the current scheme.

Conclusion Based on the results above, this research proposes a suitable design approach for sponge facility system in residential areas under different application conditions. Specifically, scheme S0 is more suitable for application in residential areas with high green land rate, low requirements for annual runoff control rate, high requirements for rainwater resource recycling, and high investment cost. S1 is suitable for application in residential areas with relatively limited land use but still with certain green space conditions, and in projects with high requirements for annual runoff control rate and tight cost input. S2 is suitable for application in residential areas with certain green space conditions, and in projects with high requirements for comprehensive hydrology-cost performance. The aforesaid design approach can well balance the requirements of hydrological control and landscape diversity in an economical and efficient manner. The results of this research have positive reference value for the construction optimization of spongy residential areas in Chongqing and other subtropical humid areas with similar precipitation conditions.