Abstract: The main urban area of Kunming City is selected as the case study site, which focuses on the spatial distribution characteristics of urban waterlogging points and the identification of homogeneous waterlogging point clusters, along with an analysis of their disaster-prone environments. Leveraging waterlogging point information publicly disclosed by news media in 2020 as the data source, the research initially employs kernel density and K-function analyses to explore the overall distribution and clustering extent of waterlogging points. Subsequently, the study utilizes the HDBSCAN spatial clustering algorithm to identify different types of waterlogging point clusters in the main urban area of Kunming City, and analyzes the characteristics of these clusters. Lastly, combined with disaster-prone environmental data, the study takes the cluster as the smallest research unit to explore the diversity of its disaster-prone environment. The research findings reveal: ① a significant spatial scale differentiation in the distribution of waterlogging points is found out, showcasing a pattern characterized by extensive dispersion and minor aggregation; ② Kunming City's waterlogging points can be divided into eight distinct types, each with a minimum of two clusters and a maximum of four; ③ Different clusters of waterlogging points exhibit some distinct disaster-prone environments among which the topographical variations have the most significant influence on the distribution of waterlogging points. In other words, the rougher and more uneven is the terrain, the higher is the likelihood of experiencing waterlogging disasters within the cluster's scope. Therefore, in urban waterlogging prevention and control efforts, differentiated mitigation measures can be formulated based on the spatial scope of clusters and their disaster-prone characteristics. This approach enables a scientifically rational allocation of mitigation efforts, thereby enhancing the scientific rigor and precision of urban waterlogging management.