Objective  As an important part of public health infrastructure and recreation service, public green space is one of the important supports for the physical and mental health of urban residents. At present, problems such as the mismatch and imbalance between the supply and demand of the recreation service of public green space are common in high-density urban areas, which leads to the insufficiency of residents’ demand for getting close to nature and outdoor recreation to a certain extent. Detailed evaluation of the recreation service capacity of public green space can better optimize recreation supply and meet public demand.

Methods  This research takes downtown Chengdu as an example, and takes the whole downtown area as the research scale. By sorting out and summarizing the theory of recreation demand, this research tries to find out what contents should be included in the supply of recreation service in a more systematic way, and then formulates an evaluation indicator system for the recreation service capacity of public green space. According to the indicator system, the research measures the individual and overall recreation service capacity of public green space, and analyzes the recreation space structure formed by public green space. In order to comprehensively understand the current supply of recreation service of public green space in downtown Chengdu, the gravity model is used to analyze the intensity of the co-opetition relationship between public green spaces.

Results  This research finds that the recreation service capacity of individual public green spaces in downtown Chengdu presents a spatial distribution feature of “lower in the center and higher on the periphery”, with the center area featuring low recreation service capacity of individual public green spaces, and the area near the Third Ring Road and its periphery featuring high recreation service capacity. The overall recreation service capacity formed by all public green spaces in downtown Chengdu presents the spatial distribution feature of “higher in the center and lower on the periphery”. Specifically, the distribution density of public green spaces within the Third Ring Road is the highest, and the overall recreation service capacity in this area is also high, which makes up for the insufficiency of recreation service capacity of individual public green space in this area. Meanwhile, the public green space outside the Third Ring Road is on a large scale and can provide high-level recreation service, but the distribution density thereof is gradually decreasing, which weakens the overall recreation service capacity of the public green space in this area. In addition, the recreation space structure of public green space in downtown Chengdu is actually a multi-mode comprehensive structure of “polar core − cluster − sector − ring belt − scatter”. The research also finds that the different co-opetition relationships between public green spaces will lead to different supply and demand models of recreation services, thus eventually affecting the efficiency of recreation services of public green spaces. In the area with strong co-opetition relationship between public green spaces, the supply and demand of recreation services are spatially decentralized. In the area with weak co-opetition relationship between public green spaces, the supply and demand of recreation services are spatially centralized. In the area with moderate co-opetition relationship between public green spaces, the public green space can provide quality recreation services for residents in and around the Third Ring Road, while supplementing daily and occasional recreation needs.

Conclusion  This research optimizes the evaluation method for the recreation service capacity of public green space from three aspects: measurement standard, research scale and cognitive object. The theory of recreation demand is innovatively introduced to assist the research in determining the evaluation indicator for the recreation service capacity of public green space. Based on the analysis results, the research proposes the following strategies: 1) Form a recreation network system featuring the trinity of “human − space − service”, optimize the facilities and functions of recreation services based on the needs of residents, and improve the quality and efficiency of recreation services through the optimization of design; 2) take the inner and peripheral parts of the downtown area as a whole to repair the dislocation of supply and demand between the inner and outer parts of the downtown area, and at the same time improve the multi-dimensional accessibility of public green space (including transportation accessibility, visual accessibility and psychological accessibility); 3) build a collaborative supply platform among multiple providers of recreation services through systematic optimization and collaborative governance, so as to solve the problem of “fragmentation” in the supply of recreation services of public green space, and effectively integrate the recreation service resources of public green space. In order to accurately identify the recreational service capacity of public green space, the factors such as demand intensity and distance barrier is not taken into account in this research, which weakens the correlation effect between supply and demand factors. Future researches are suggested to, based on the availability of data, further improve the evaluation indicator system or evaluate the actual benefit of the recreational spatial structure.