Spatial-temporal Differentiation and Influencing Mechanism of Accessibility of Park Green Spaces: A Case Study Based on Shenzhen City

Abstract

Park green space is an important part of the urban natural ecosystem and public service facilities. The accessibility index of park green space can characterize the difficulty of using park green space for residents. In addition, it is an important index that reflects the rational layout and fair use of park green spaces. With the rapid development of urbanization, the unrestrained expansion of construction land, and the continuously increasing demands of the population have led to a gradual imbalance between the supply and demand of park green spaces. In this study, the spatialtemporal differentiation characteristics of accessibility of multi-scale park green spaces in Shenzhen under multiple travel modes from 2013 to 2020 were analyzed using the Gaussian two-step floating catchment area (Ga2SFCA) method, which comprehensively considers both supply and demand. The influencing degree of each factor on the accessibility of park green space was studied with a geographical detector from the global scale, through which the dominant factors were screened. The spatial differentiation of the action intensity of each dominant factor on the local scale was investigated using the geographically weighted regression (GWR) model. Results showed that from 2013 to 2020: 1) Within the scope of construction land, the number and area of green parks in Shenzhen increased continuously. The number of community parks increased the greatest amount, and the area of comprehensive parks expanded the greatest amount. There was an imbalanced distribution pattern, which is characteristic of “more in the west and less in the east”,“from the inner areas to the outer areas”, and expansion from south to north. 2) The demand of residents for park green space was positively correlated with population density. The population density in the western region of Shenzhen was increasing gradually, and the population density in the eastern region was always low. The high-value area of green space demand extended gradually from the inner areas to the outer areas. 3) The accessibility of park green space generally presented a rising trend. The proportion of high-value residential areas was increasing gradually. The accessibility of the outside area became gradually higher than that of the inside area. The average accessibility of community parks achieved the highest growth rate, while the accessibility of comprehensive parks was basically stable. The standard deviation of special parks increased the most, and the spatial differentiation was intensified. 4) In 2013, the p-value of all influencing factors was 0, which passed the 0.01 significance test. In 2020, only the three factors of park area proportion, park entrance and exit density, and road density passed the 0.01 significance test, indicating that the spatial distribution of park accessibility from 2013 to 2020 was closely related to green space factors and road factors. The correlations of park accessibility with location, economy, and natural factors presented a downward trend with the development of science and economy. 5) According to the determination power (q-value) of the dominant factor, the q-values of the proportion of park area, park entrance and exit density, and road density all showed a downward trend on the dynamic level. The factors can be ranked road density > proportion of park area > park entrance and exit density in terms of action intensity. This revealed that road density was a dominant factor in park accessibility, with an explanatory power of as high as 10% , which was much larger than that of other factors. The park entrance and exit density and the proportion of park area were important factors for park accessibility. 6) The influences of different factors on the accessibility index of park green space have significant spatial differences. Road density (X4) and proportion of park area (X1) had a greater impact in the eastern study area, while park entrance and exit density (X2) had a greater impact in the western study area. The research enriches perspectives on green space accessibility. The method of exploring the spatial-temporal differentiation characteristics and influencing mechanism of green space accessibility not only can be applied to the layout and planning of urban park green spaces, but also provides a reference to analyze other forms of geographical accessibility.