Exploring the correlation of the built environment with metro ridership is vital for fostering sustainable urban growth. Although the research conducted in the past has explored how ridership is nonlinearly influenced by the built environment, less research has focused on the spatiotemporal ramifications of these nonlinear effects. In this study, density, diversity, distance, destination, and design parameters are utilized to depict the “5D” traits of the built environment, while Shapley Additive Explanations with eXtreme Gradient Boosting (XGBoost-SHAP) are adopted to uncover the spatial and temporal features concerning the nonlinear relationship of the built environment with ridership for metro stations located in Xi’an. We conducted a K-means clustering analysis to detect different site clusters by utilizing local SHAP coefficients. The results show that (1) built environment variables significantly influence metro ridership in a nonlinear manner at different periods and thresholds, with the POI facility density being the most critical variable and the other variables demonstrating time-driven effects; (2) the variables of population density and parking lot density exhibit spatial impact heterogeneity, while the number of parks and squares do not present a clear pattern; and (3) based on the clustering results, the metro stations are divided into four categories, and differentiated guidance strategies and planning objectives are proposed. Moreover, the current work offers a more developed insight into the spatiotemporal influence of built environments on metro travel in Xi’an, China, using nonlinear modeling, which has vital implications for coordinated urban–metro development.

With the operation of the first route in Xi’an City, the matching relationship between the metro networks and the urban population is a root factor affecting the utilization of rail transit facilities. The mismatch between the metro networks and the urban population has led to an imbalance between the supply and demand for rail transport, resulting in wasted urban infrastructure. Based on this issue, the research objective is to focus on the spatiotemporal variations of the matching relationship. Firstly, the topological network model abstractly extracted metro spatial distribution features, and the spatial autocorrelation model was adopted to identify the evolution characteristics of the metro networks and urban population. Secondly, this paper adopted a time-lagged regression model to demonstrate the action relationship from 2011 to 2021. Then, the compositive coordination index was utilized to assess the variation of the global matching relationship. Finally, the paper explored spatial heterogeneity through the coupling coherence degree attached to grid cells. The research results indicate that the Moran’s I value of metro elements decreased from 0.782 to 0.510 with the further complexity of topological networks, while the population was consistently high in spatial dependence with a Moran’s I value of around 0.75 during the decade. Based on the regression coefficients and significance, this paper verified the hypothesis that the metro networks and urban population had a positive time-lagged feedback effect in urban development. From 2011 to 2021, the compositive coordination index symbolizing the global matching relationship increased from 0.29 to 0.90, but the coupling coherence degree shows significant spatial heterogeneity in different grid units. Differentiated spatial planning strategies were proposed for varied areas to efficiently utilize rail transit, which may provide a reference for other cities with the same reality problem.

In recent years, Xi'an metro construction has been progressing rapidly, becoming the primary mode of urban green public transportation. Since the ridership of the metro is closely linked to the characteristics of its surrounding built environment, a key problem in promoting the benign development between the two is to explore the spatiotemporal distributional difference in ridership and its influencing factors. In this study, the "5D" characteristics of built environment are described by density, diversity, design, destination and distance variables. The spatiotemporal distribution characteristics of ridership are analyzed via Arc GIS and Python, while the nonlinear relationships between ridership and built environment of 106 metro stations of downtown Xi'an, as well as relevant threshold effects are revealed via Shapley additive explanations with gradient boosted decision tree (GBDT-SHAP). The results show that: (1) Xi'an metro travel presents a medium-short spatiotemporal distribution, and the ridership network is characterized by strong center-spillover. (2) The nonlinear relationship between built environment and ridership is ubiquitous and presents a threshold effect. The impact threshold of bus stop density on ridership is 4-6 pcs/km², the impact threshold of road network density is roughly 4-5 km/km², and the effective threshold of building density does not exceed 20%. (3) The positive impact of POI facility density on peak ridership is stronger than that at flat hours. Variables like land use mixture, population density and distance from downtown have a time-driven effect on the ridership, whose importance and influence change with time. This study provides a better understanding of the spatiotemporal impact of Xi'an's built environment on metro travel, which is of profound significance for the coordinated development between the city and metro construction.