The rapid development of the city leads to the continuous updating of the ratio of land use allocation, especially during the ood season, which will exacerbate the signi cant changes in the spatial and temporal patterns of urban ooding, increasing the di culty of urban ood forecasting and early warning. In this study, the spatial and temporal evolution of ooding in a high-density urban area was analyzed based on the Mike Flood model, and the in uence mechanisms of different rainfall peak locations and in ltration rate scenarios on the spatial and temporal characteristics of urban waterlogging were explored. The results revealed that under the same return period, the larger the rainfall peak coe cient, the larger the peak value of inundation volume and inundation area. When the rainfall peak coe cient is small, the higher the return period is, and the larger the peak lag time of the inundation volume is, in which P = 50a, r = 0.2, the delay time of the inundation volume for the inundation depths H > 0.03 m and H > 0.15 m reached 32 min and 45 min, respectively, At the same time, there are also signi cant differences in the peak lag time of waterlogging inundation volume in different inundation depths. The greater the inundation depth, the longer the peak lag time of waterlogging inundation volume, and the higher the return period, the more signi cant the effect of lag time prolongation. It is worth noting that the increase in in ltration rate will lead to the advance of the peak time of inundation volume and inundation area, and the peak time of the inundation area is overall more obvious than that of inundation volume. The peak times of inundation volume and inundation area were advanced by 4 ~ 8 min and − 2 ~ 9 min for H > 0.03 m and H > 0.15 m, respectively, after the increase in in ltration rate; and the higher the return period, the smaller the rainfall peak coe cient and the longer the advance time. The spatial and temporal characteristics of waterlogging under different peak rainfall locations and in ltration capacities obtained in this study can help provide a new perspective for temporal forecasting and warning of urban waterlogging.