Abstract: Short-duration heavy precipitation is a primary hazardous weather event during the rainy season in Kunming. Conducting an in-depth study of its spatiotemporal characteristics and influencing systems is of great significance for enhancing local forecasting and warning capabilities for heavy precipitation. Based on hourly precipitation data from 12 national meteorological stations (referred to as national stations) and 406 high-density regional automatic meteorological stations (referred to as regional stations) in Kunming from 2018 to 2022, this study employed statistical and synoptic diagnostic methods to analyze the spatiotemporal distribution characteristics of short-duration heavy precipitation during the rainy season (May to October) and performed circulation classification for regional events. The results show that spatially, short-duration heavy precipitation exhibited a pattern of "more frequent in the south and less in the north," with the main urban area demonstrating a "rain island effect" characterized by the highest frequency and intensity. Temporally, its occurrence was most concentrated during the main flood season (June to August), with the peak of ordinary short-duration heavy precipitation in July, while the peak of extreme short-duration heavy precipitation was in June. The diurnal variation presents a bimodal structure, with primary peaks in the early morning and late afternoon. The main weather systems influencing regional short-duration heavy precipitation events in Kunming are classified into four types: shear line type (50.7%), tropical low-pressure system type (20.8%), two-high convergence type (17.4%), and subtropical high periphery type (11.1%). Differences in moisture source, topographic coupling, and dynamic configuration are the primary causes for the distinct precipitation characteristics of different systems. The surface convergence line (or front) combined with the 700 hPa shear line is the key dynamic combination for triggering short-duration heavy precipitation, providing valuable guidance for forecasting.