In response to the issue of how changes in rainfall patterns under the context of climate change affect flood disaster processes, a oneway coupled urban pluvial flood model based on SWMM and LISFLOODFP for the independent drainage area of Jinzhou Creek in Nansha, Guangzhou was constructed. The flood characteristics under storm conditions with peak rainfall coefficients of 0.431 (standard rainfall pattern), 0.2 (rainfall pattern Ⅰ), 0.5 (rainfall pattern Ⅱ), and 0.8 (rainfall pattern Ⅲ) were examined. The results show that under the standard rainfall pattern design storm, as the rainfall return period increases from 2 a to 100 a, the maximum flow at the overflow node increases by about 55%, the total overflow volume rises by 207%, and the inundation area proportion expands from 24.2% to 40.8%. As the rainfall return period increases, the area of severe inundation increases significantly. As the peak rainfall coefficient increases, the time to reach the peak flow is delayed, and the lag time to the maximum inundation area becomes shorter. The peak time for inundation area occurs earlier than the peak time for total surface water accumulation, with the earlier the peak time for inundation area being, the smaller the peak rainfall coefficient. Under different rainfall pattern design storm conditions, the inundation area is positively correlated with the peak rainfall coefficient, and the growth rate of inundation area is negatively correlated with the rainfall return period. The symmetric rainfall pattern results in the largest total overflow, with the overflow volume under rainfall pattern Ⅱ being, on average, 3.765% higher than the standard rainfall pattern, and it can be as much as 7.728% higher at maximum. As the rainfall return period increases from 2 a to 100 a, the time at which the pipeline network reaches overload and begins to overflow occurs 15 to 40 min earlier, with both the overload and overflow durations being extended.