［Objective］ The spatiotemporal evolution characteristics and influencing factors of flood disasters in Pakistan were analyzed， in order to provide scientific references for flood risk analysis and management. ［Methods］ Based on 143 typical flood events recorded by the Dartmouth Flood Observatory （DFO） from 1985 to 2021， this study integrated multi-source spatial datasets， including ERA5-Land precipitation， digital elevation model （DEM）， land use， population density， and gross domestic product （GDP）. The temporal variation， spatial distribution， gravity center migration， and directional evolution of flood disasters were analyzed using methods such as anomaly analysis， the gravity center model， and the standard deviational ellipse. The optimal parameter-based geodetector （OPGD） model was employed to quantitatively identify the dominant influencing factors and their interactive effects on the spatiotemporal distribution of flood disasters. ［Results］ Flood disasters in Pakistan entered a period of high occurrence from 2005 to 2010， with the frequency reaching a historical record high in 2007. The disasters exhibited spatial evolutionary characteristics， shifting from widespread distribution to high-density and high-severity concentration in the middle and upper reaches of the Indus River. The disaster centroid migrated northward， the primary distribution axis aligned in a north-south direction， and the spatial pattern became increasingly clustered. The geodetector analysis revealed that precipitation-related factors （Rx_5d， R_20mm， and Rx_1d） possessed strong explanatory power， with most combinations of factors exhibiting significant synergistic effects. ［Conclusion］ Flood disaster risks in Pakistan steadily increased， characterized by shifting disaster centroid and distribution orientation， while the spatial pattern becomes increasingly concentrated. Extreme precipitation events serve as the key trigger of flood disasters， while the interaction between topographic factors and precipitation-related factors critically shapes the spatial heterogeneity of flood occurrence.