Multi-method combined analysis of urban flood risks and its influencing factors under low impact development

In previous studies, the mitigation of urban floods via low impact development (LID) has typically been expressed only in terms of runoff volume reduction. The impact of duration time and flow velocity on the flood risk has not been quantitatively analyzed. This paper takes the Beiyangyuan Campus of Tianjin University (national pilot sponge campus) as the study area. Based on the constructed SWMM and MIKE21 coupled model, the effects of LID on different flood characteristics are analyzed by setting LID removal scenarios. The results show that LID mainly mitigates urban floods by reducing the runoff volume, inundation area and risk area, and the reduction effect is approximately 15%–38%. The water depth, flow velocity and duration time are not sensitive to LID practices, and the reduction effect is basically below 10%. The flood risk and its influencing factors in the study area were analyzed using combined multi-methods based on water depth (WD), water depth & duration time (WD&DT), and hazard rating (HR). The results show that the central island with numerous LID practices has an almost undetected flood risk, and the high risk is mainly distributed in the northeast of the study area. Duration time is an important influencing factor on flood risk, and after adding consideration of duration time, the proportion of risk levels upgraded to moderate or high is more than 90%. Further analysis shows that insufficient drainage capacity of the pipe network leads to prolonged ponding, which is the primary cause of flood risk in the study area. Adding the consideration of flow velocity, the 27–36% low-risk area changes to moderate risk; multiple linear regression analysis shows that the contribution of flow velocity to the flood risk is minimal. The possibility of flow impact hazard in the study area is very low, and the cumulative frequency of water depth and flow velocity product (d·v) less than 0.04 m2/s is more than 98% in each scenario.