AbstractUrban rail transit has effectively alleviated the pressure on road traffic. To explore the key influence factors that a rail station’s built environment has on passenger flow and its heterogeneity along with temporal and spatial changes, in this paper, a geographically and temporally weighted regression (GTWR) model was constructed to identify. Specifically, an empirical study was conducted in Xi’an, China, using 1 month of smartcard and station-level point-of-interest data. Firstly, we extracted an influence factors set (IFS) for ridership at the station level, and thereby three aspects of characteristics were obtained to establish IFS, including land usage, interchange connection facilities, and attributes for the station. Then, variables were exactly determined to describe each aspect characteristic with the analysis of the multicollinearity and spatial self-correlation. In addition, for models, ordinary least squares (OLS), geographically weighted regression (GWR), and geographically and temporally weighted regression (GTWR) were built to explore variables’ heterogeneity and variation influence in spatiotemporal for station ridership over time and location. Results reveal that GTWR outperforms in effectively capturing the spatiotemporal performance of ridership influences. Moreover, we proposed a mixture Poisson model to cluster stations with typical land-use characteristics in order of GTWRs’ application in different types of stations, for practice. In sum, ridership changes of different stations affected by a specific influences over time were analyzed, which highlighted the importance of temporal features in spatiotemporal data. Using GTWR to explore the relationship between ridership and station environment can provide insightful essential information for policymaking in urban rail transit management.