How localized differences in street greenery conditions are associated with urban public space use patterns: Evidence from the Las Vegas Strip

In hot-arid cities, street-level greenery is often introduced with the expectation that it will improve pedestrian comfort, for example by providing shade and moderating local thermal conditions. However, it remains unclear whether street-level greenery meaningfully shapes pedestrian behavior in places where foot traffic is already intense and urban functions are highly concentrated. To address this question, we compare street greenery and hardscape segments along the Las Vegas Strip, one of the world’s most heavily used pedestrian corridors, using commercially collected, anonymized mobile location data from 2020 to 2024. Our analysis proceeds in three connected steps, moving from temporal patterns of pedestrian activity to movement sequences, and finally to identifying locations with high potential for future greening interventions. First, we identify pedestrian stops based on dwell time and summarize pedestrian activity over time to characterize patterns of visitation, time spent, and hourly use. Group comparisons and temporal clustering show that areas with street greenery do not consistently attract more pedestrians overall, but their use shifts over the course of the day, with a larger share of activity occurring at night. Second, reconstructing movement paths and examining the network structure suggests that these behavioral differences do not meaningfully change overall movement patterns along the corridor. Instead, green sites tend to play different directional roles within pedestrian sequences, more often serving as points of origin than destinations, especially at night. Finally, we use supervised learning to identify locations with high potential for future street greenery, drawing on temporal, spatial, and contextual patterns in pedestrian activity. The models highlight spatially coherent areas with higher intervention potential, illustrating how observed pedestrian behavior can be used for exploratory screening of candidate sites. The findings indicate that contrasting localized streetscape conditions are most consistently associated with differences in nighttime activity shares and localized lingering behavior, rather than with higher aggregate pedestrian volume or changes in corridor-wide movement structure. More broadly, the study demonstrates how multi-year mobile location data can support descriptive behavioral mapping and exploratory screening of candidate streetscape improvement zones under shared climatic variability, rather than causal evaluation of greenery impacts.