Effect of Working Patterns on Spatial Variation in Outdoor Leisure Activity of Fixed-Location Workers: A Case Study of Nanjing, China

A comprehensive understanding of intra-urban dynamic changes requires an in-depth analysis of outdoor leisure activity behaviors. However, the existing research has not systematically examined how working hours interact with spatial factors in shaping leisure behavior. To bridge this gap, we investigated the spatial differentiation of outdoor leisure activities on weekends among fixed-location workers in Nanjing and evaluated the influence of working patterns. The study area consists of 14 urban zones, including two main cities, three sub-cities, and nine new towns. First, behavioral features were extracted from mobile phone data. Second, an autoencoder was used to construct a weekend outdoor leisure activity indicator (Participation). Lastly, the Geodetector method quantified the independent effects of the working patterns, accessibility features, and individual features. In addition, we examined the spatial heterogeneity of these factors and determined how their interactions enhance their overall impact. The results indicate that, in 12 urban zones, variations in Participation are primarily shaped by working patterns, with the explanatory power (q) ≥ 0.04. The dominant factors exhibited a layered distribution, expanding outward from the center of Nanjing in a linear pattern, with an average expansion interval of 8.9 km and a coefficient of determination (R²) of 0.996. Moreover, interactions among different factors substantially amplified their effects, with q increases as high as 257.1%. These findings underscore the pivotal role of working patterns in shaping residents’ outdoor leisure activities on weekends and highlight the spatial differences in the driving factors, offering new insights into the spatial and social trends of leisure behavior.