<p>Green roofs have been shown to significantly mitigate urban heat island effects and serve as an important solution for addressing the shortage in urban green spaces. Their capacity to improve the surface thermal environment is influenced by both the morphology of the urban blocks and their landscape patterns. In this study, surface temperatures for four seasons (spring, summer, autumn, and winter) were retrieved, and the morphological types of the central area of Guangzhou were classified using the local climate zone (LCZ) framework. Vector data on green roofs were collected to construct a cooling model and calculate landscape pattern metrics. A geodetector model was employed to analyze the cooling mechanisms of the green roofs. The results indicated that the cooling effect of green roofs is most pronounced in summer, with the highest average cooling intensity of an area reaching 0.63℃. The optimal LCZ type for green roof cooling performance was found to be that of compact low-rise buildings, and compact building areas were identified as more suitable than open building areas for green roof implementation. Furthermore, larger average green roof areas were associated with enhanced cooling effects.</p>