Effects of neighborhood parks on urban vibrancy in metropolitan Tokyo using propensity score matching and machine learning

Abstract In recent years, many metropolitan areas have faced declining opportunities for social interaction and reduced local vitality owing to changes in population structure and the uneven distribution of urban functions. Accordingly, interest in the role of public spaces has increased, particularly in relation to the creation of walkable urban environments. Urban parks may support local vibrancy by encouraging everyday staying activities, pedestrian movement, and social interaction. However, empirical evidence on the relationship between neighborhood park presence and multiple urban activity indicators remains limited, particularly after accounting for differences in surrounding urban conditions. This study examines the relationship between neighborhood park presence and urban vibrancy indicators within station catchment areas in the Tokyo metropolitan region. Station catchment areas are defined as walkable zones surrounding railway stations that reflect daily activity spaces. Pedestrian flow derived from mobile phone location data, child population, and annual retail sales were used as indicators representing different dimensions of urban activity. Propensity score-based methods were applied to compare station catchment areas with and without neighborhood parks while adjusting for observed differences in facility distribution, land use, population characteristics, and transportation conditions. The results show that neighborhood park presence is positively associated with child population and weekday pedestrian flow among workers, whereas no clear associations were observed for annual retail sales or most other pedestrian-flow indicators. These findings suggest that neighborhood parks may be related to family-oriented residential environments and weekday worker mobility within station catchment areas. However, because the analysis is based on observational and cross-sectional data, the results should be interpreted as adjusted associations rather than definitive causal effects of park development.