PROGRESS IN GEOGRAPHY ›› 2023, Vol. 42 ›› Issue (4): 716-729.doi: 10.18306/dlkxjz.2023.04.008

• Articles • Previous Articles     Next Articles

Nonlinear relationship between urban vitality and the built environment based on multi-source data: A case study of the main urban area of Wuhan City at the weekend

WANG Zimeng(), LIU Yanfang*(), LUO Xuan, TONG Zhaomin, AN Rui   

  1. School of Resource and Environmental Science, Wuhan University, Wuhan 430079, China
  • Received:2022-11-02 Revised:2023-02-24 Online:2023-04-28 Published:2023-04-27
  • Contact: LIU Yanfang;
  • Supported by:
    Key Program of National Natural Science Foundation of China(42230107)


Constructing a vibrant urban space by optimizing urban built environment is a significant measure to shape and develop urban vitality. However, the nonlinear and threshold effects of variables on urban dynamics are often overlooked in most existing studies. In this study, we used the Wuhan metropolitan area at the weekend as a case to quantify the spatiotemporal pattern of urban vitality with population heat data in 2019. We investigated the nonlinear influence of urban built environment on urban vitality and extracted the optimal thresholds for key variables with the gradient boosting decision tree-Shapley additive explanations (GBDT-SHAP) model. The results indicate that: 1) The distribution of urban vitality showed distinct spatiotemporal and functional heterogeneity in the Wuhan metropolitan area. Spatially, urban vitality showed an overall characteristic of high in the center and low at the periphery; temporally, it gradually increased from 7:00 a.m. and peaked in the afternoon and evening; functionally, urban vitality was higher in areas with better infrastructure facilities and near large shopping centers, and lower in industrial areas. 2) The built environment in the Wuhan metropolitan area significantly affected the intensity of urban vitality, and its nonlinear influences and threshold effects were significant: the impact threshold for large shopping centers was roughly 3000-4000 m; the impact of subway on urban vitality was mainly in the intra-city area and the subway stations served the areas within 1500 m; the point of interest (POI) mix above 0.4 inhibited urban vitality. The SHAP values compensated for the shortcomings of traditional multivariate linear models in terms of interpretability. 3) The three variables of distance to central business district (CBD), distance to subway stations, and sky openness and business type have time-varied effects, and the importance of their impacts on urban vitality changes over time. Implementing the transit-oriented development (TOD) concept and improving the connection of metro traffic with large shopping areas are effective measures to enhance the vitality of the city. Setting up outdoor open space promotes residents’ morning exercise and social activities. Adding small dining and recreational spaces is also a measure to enhance vitality. These nonlinear effects and thresholds help planners make better decisions.

Key words: urban vitality, urban built environment, nonlinear relationship, threshold effect, GBDT-SHAP, Wuhan City