The service level and fairness measurement of urban park green space is one of the research hotspots in the field of environmental justice. Under the background of urban low-carbon transformation and sustainable development, the theory of landscape ecology is introduced to construct the calculation model of ecological supply index of park green space. Secondly, the demand index system is constructed by using POI, second-hand housing and other spatio-temporal big data with the intensity of human activities as the center. Finally, the Gini coefficient and location entropy are used to measure the fairness of supply and demand balance, and the supply and demand types are divided based on the Z-score method. Taking Shenzhen as an example, the results show that the city 's global park green space resources are rich, but the Gini coefficient of 0.489 indicates that the fairness is poor, and there is social differentiation in spatial distribution. The specific performance is as follows: 1) The location entropy value shows a spatial pattern of high in the east and low in the west. 2) The high value area of location entropy is mainly distributed in Dapeng New District, which is rich in green space resources, and the supply and demand types are mainly high supply and low demand. The low-value areas are mainly distributed in industrial agglomeration areas such as Bao'an West and Longgang West, with low supply and high demand. 3) The resource allocation of park green space is polarized. A small number of areas occupy most of the resources, while the resources of the burden areas are insufficient. According to the experimental results, this paper puts forward some suggestions on the planning of urban park green space, hoping to further promote the balanced development of urban park construction and green space.