Abstract: Convection Initiation (CI) is an critical precursor to severe convective weather, and its accurate identification and early warning are of great significance for meteorological disaster prevention and mitigation. Using high spatiotemporal resolution radar composite reflectivity mosaic data and surface automatic weather station precipitation data from Hubei Province during 2020–2024, this study applied an improved high-resolution grid-based objective CI identification algorithm to construct a CI dataset. The spatiotemporal distribution characteristics of CI and their relationship with topography were subsequently analyzed. The results are as follows. (1) A total of 107,322 CI number were identified in Hubei Province from 2020 to 2024, corresponding to an average of 59 events per day and a CI-day occurrence frequency of 73%. Both the daily mean CI number and the frequency of CI days were highest in 2020 and lowest in 2022. Seasonally, CI activity was most active in summer, followed by autumn and spring, and least active in winter. (2) The monthly variation of CI activity exhibited a bimodal pattern, with the primary peak observed in June–July, corresponding to the Meiyu season, and a secondary peak noted in October–November, associated with seasonal transition, cold-warm air convergence, and the influence of tropical systems such as typhoons. (3) The diurnal variation of CI was characterized by a single or dominant peak from afternoon to evening, with a secondary peak identified in the late night during summer. Except in autumn, daytime CI occurrences were generally higher than those at night. (4) The distribution of CI was found to be uneven and significantly influenced by underlying surfaces. High-frequency areas were concentrated mainly in the mountainous regions of southwestern Hubei, particularly along the Enshi-Yichang border. Other relatively active areas included the Mufu Mountains in southeastern Hubei, the Dabie Mountains in northeastern Hubei, mountainous regions in northwestern Hubei, and areas surrounding large lakes and reservoirs. In contrast, the central plains and basins exhibited relatively lower CI frequencies. These findings provide a preliminary understanding of the occurrence and evolution characteristics of CI activity in Hubei Province and offer a scientific basis for the development of a localized CI forecasting model. Such models can improve the early detection and warning of severe convective weather, thereby enhancing meteorological disaster preparedness and mitigation efforts.