Visitation-based classification of urban parks through mobile phone big data in Tokyo
Yichun Zhou,ChengHe Guan,Longfeng Wu,Ying Li,Xuanyi Nie,Jihoon Song,Seung Kyum Kim,Yuki Akiyama
DOI: https://doi.org/10.1016/j.apgeog.2024.103300
IF: 4.732
2024-05-22
Applied Geography
Abstract:Urban parks, pivotal in fostering physical activity, mental well-being, and environmental stewardship, are integral to green infrastructure planning. Despite advances in georeferenced data applications, existing park classifications often overlook actual visitation patterns. This study reclassifies urban parks using over 5.9 million records from approximately 330 thousand visitors across 300 Tokyo parks, comparing with size-based park categorizations. We employed a range of analytical tools, including principal component analysis, Isolation Forest algorithm, various clustering algorithms, and the Gini index. Our findings first revealed four key visitation indicators, activity intensity, utilization efficiency, temporal occupancy, and revisit volume. These indicators uncovered parks with atypical visitation patterns within each size category, leading to three novel park classifications, everyday leisure parks, social destination parks, and seasonal activity parks. Moreover, we discovered notable disparities in distances traveled to parks, particularly during nights, weekends, and holidays, with pronounced inequalities in seasonal activity parks and smaller parks. The findings advocate for a nuanced park management strategy, prioritizing maintenance and amenity development aligned with observed visitation patterns to enhance recreational potential, thereby contributing insights to urban park research that support the advancement of green infrastructure planning and policy aimed at improving park utility and enjoyment.
geography