It is critical to accurately assess the location and evolution of impending landslide hazards, especially when influenced by the effects of climate change. However, most existing landslide models only analyze the stability of individual grid cells without considering soil interaction between adjacent cells. This means that the stability of individual grid cells is not necessarily representative of actual slope stability at a given location, in turn affecting the accuracy of landslide stability assessments. In this paper, slope units are defined using the morphological image analysis and homogeneous slope unit method to represent the actual slope conditions and to extract the profile of each slope unit. The critical slip surface of the soil mass is investigated to account for the interaction between adjacent grid cells within the slope unit. A probabilistic analytical landslide model is established and the response mechanism of the hydrological processes is proposed. This model is applied to landslide hazard assessment in the Three-Gorges Reservoir Area, Fengjie County, China. Using historical rainfall data, the distribution of landslide hazards is analyzed for different antecedent precipitation levels and rainfall return periods. The results show that the amount of antecedent precipitation, and rainfall return period, have different effects on landslide triggering in the area. Importantly, the model reveals that extreme rainfall events induced by climate change will lead to a large number of new landslides in areas where they are not anticipated. Further investigation and monitoring (e.g., early-warning systems, long-term assessment) of rainfall-induced landslides in the Three-Gorges Reservoir Area is recommended.