Abstract: At present, the rapid urbanization construction has led to a series of ecological environment problems, such as green space shrinkage and ecosystem degradation. The construction of “green belt area” in the urban fringe is an important means to effectively guide the rational layout of urban greening and meet the needs of the capital in ecological and environmental protection. Taking the second green belt area of Beijing as an example, this research applies the Fragstats model, CASA model, InVEST model and multi-source big data to evaluate the spatial and temporal evolution characteristics of regional landscape pattern diversity index and ecosystem support services. It also adopts the GWR model to explore the driving and feedback mechanisms between the two. The results demonstrate a decreasing trend of the landscape diversity of the second green belt area from 2003 to 2018. The support services displayed certain characteristics of circle spatial distribution changes. Besides, the driving effect of the support services provided by the ecosystem on the landscape diversity is far greater than the impact of the changes of landscape diversity on the support services. The innovation of the research lies in the application of multi-source big data and spatial analysis model to quantitatively evaluate the spatial-temporal variation characteristics of landscape diversity and ecosystem support services in the second green belt area, and the exploration of the interaction mechanism between them combined with the spatial regression model. This research is conducive to scientifically identify the areas with strong correlation between ecosystem support services and landscape diversity, and could provide scientific guidance for the planning and construction of the subsequent green belt areas and the promulgation of various policies.