To investigate the spatial behavioral manifestations of heterogeneity in tourists’ travel durations, this study selects the central urban area of Chongqing ( a popular tourist destination in China) as the research region. Utilizing social network analysis and GIS spatial analysis, we explore the structural characteristics of tourist flow networks under different travel durations. The results reveal that: Prolonged travel durations drive the outward expansion of tourist flows, enhancing the diversity of attraction selection in peripheral areas. Spatial tourist flows across all durations exhibit an exponential distance-decay effect, most pronounced in two-day trips, followed by oneday trips, and weakest in trips lasting three days or longer. Extended durations correlate with increasingly complex network structures, where network density, connectivity, and structural hole advantages increase correspondingly. Core attractions maintain strong-to-strong connections regardless of duration, while path multiplicity and diversified route combinations emerge with longer stays. A reinforced core-periphery structure emerges as durations extend, demonstrating limited spillover effects from core to peripheral areas. These findings provide theoretical support for optimizing spatiotemporal resource allocation in mountainous cities.