Refining the accessibility evaluation of urban green spaces with multiple sources of mobility data: A case study in Shenzhen, China

Abstract

Urban green space (UGS) is an important component of urban resources which contributes to human physical and mental health. Studies on the accessibility of UGS under the two-step floating catchment (2SFCA) framework have recently received much attention. However, the effects of people’s actual mobility patterns have not been fully considered in current studies. Proposed in this study is an improved accessibility model called AM-Ga2SFCA, which refines the traditional Gaussian 2SFCA method with the actual mobility information extracted from mobile phone big data and online map. A new attractiveness index of UGS is implemented by combining the popularity evaluated by the PageRank algorithm and the actual utilisation based on buffer analysis. In addition, realistic travel time between each demand point and UGS is retrieved from the online map, which is further introduced into AM-Ga2SFCA as the travel cost. A case study is conducted in Shenzhen, China to validate the proposed model. Results show that the accessibility of UGS is strongly correlated with regional urbanization level, for example, higher accessibility generally occurred in the region with developed transportation and rich green resources. From the perspective of age groups and travel modes, we found that the environmental justice issue had already occurred in Shenzhen especially for the non-elderly: under the walk mode, nearly 80% of the non-elderly only shared 20% of UGS whilst approximately 80% of the elderly shared 30% of UGS. However, taxi or private vehicles can effectively alleviate the aforementioned phenomenon by reducing the Gini index to less than 0.5. The proposed model is expected to advance the understanding of UGS accessibility and facilitate effective planning to reduce environmental justice.