As China’s population continues to age, the renovation of min urban green spaces to be age-friendly has emerged as a vital concern for promoting the health and well-being of elderly residents. This research examines 10 mini urban green spaces in Nanjing and constructs a simulation model using multi-source data to elucidate the combined eff ects of spatial confi guration and vegetation morphology on older adults’ activities. The fi ndings indicate that: (1) The three-dimensional green view index (GVI3D) is a core regulatory factor, exhibiting a signifi cant threshold eff ect—when GVI3D＜0.06, spatial layout dominates activity distribution, whereas when GVI3D≥0.06, vegetation characteristics (e.g., layering, compactness) become key drivers of high-intensity activities; (2) GVI3D and spatial confi guration exhibit nonlinear interactions, jointly shaping activity typology and spatial diff erentiation. The study proposes a synergistic optimization strategy for older adults’ activities and green space design: in low-GVI3D areas, priority should be given to optimizing spatial layout and rest facilities, while in high-GVI3D areas, vegetation layering should be enhanced to guide activity patterns. The developed simulation model provides methodological support for precise age-friendly green space design, off ering practical value for promoting elderly-friendly urban development.