Tropical Geography ›› 2023, Vol. 43 ›› Issue (5): 872-884.doi: 10.13284/j.cnki.rddl.003682

Previous Articles     Next Articles

Spatial Heterogeneity of the Built Environment Effect on the Use of a Bikeshare-Metro Commute in a Metropolitan Area: A Case Study of Shenzhen

Yuanyuan Guo1(), Lei Wu2, Peng Zeng1()   

  1. 1.School of Architecture, Tianjin University, Tianjin 300072, China
    2.School of Architecture and Planning, Hunan University, Changsha 410082, China
  • Received:2022-05-27 Revised:2022-07-27 Online:2023-05-05 Published:2023-06-13
  • Contact: Peng Zeng;


Recently, dockless bikeshare (DBS), a new bikeshare program that does not feature fixed dock stations, has been developed rapidly in China and has also offered a decent solution to the first- and last-mile problem. In addition, the integration of DBS and metro strongly promotes the transit-oriented development, particularly in the metropolitan areas of China. To achieve the seamless connection between DBS and metro transit, the spatial variation of the effects of urban built environment, particularly in high-density metropolitan areas, should be explored to advance the targeted policy interventions in different urban spaces. Using data from one of the largest DBS operators in China (ofo), this study measured the integrated use of DBS and the metro quantitatively, and it employed geographically and semiparametric geographically weighted regression (GWR and SGWR, respectively) to examine the effects of the built environment on the integrated use, using Shenzhen as a case study. The findings show that (1) The SGWR model performs better than GWR and OLS in explaining the relationship between built environment and access integrated use, whereas SGWR is not applicable for the egress integrated use. (2) Three positive determinants of population density, major road length and inbound metro ridership, and three negative determinants of metro density, secondary school, and intersection density have been examined with spatial effects on the access integration. (3) For egress integration, official land use, park, bus stops, and outbound metro ridership have positive and spatial effects, while residential land use, number of secondary school, metro density and distance to CBD (Central Business District) exert negative spatial effects. The results indicate that the built environment elements usually affect the integrated use with spatial variation. Furthermore, the access and egress integration use of DBS metro largely depends on the characteristics of built environment of the origin and destination metro catchment, respectively. This work provides insight into how the DBS-metro integration, which is divided into access and egress patterns, is spatially affected by urban built environment in the Chinese metropolitan context. The results will also provide a reference for the local government to carry out the targeted policies and planning to encourage the connection between DBS and metro transit more successfully. For DBS operators, the results also contribute to allocating the bikes more efficiently, which is adapted to the dynamic demand-supply at different urban spaces.

Key words: bikeshare, built environment, Semi-Parametric Geographically Weighted Regression (SGWR), transfer behavior, spatial heterogeneity, Shenzhen City

CLC Number: 

  • U491.225