Informing inclusive park planning: Neighborhood park visitation modeling based on smartphone big data in Austin, Texas

This research uses smartphone-based big data to explore how socioeconomic, built environment characteristics, and spatiotemporal factors influence neighborhood park use, with a specific focus on minority groups’ park visitation patterns. Panel data are collected through SafeGraph, and combined with GIS and American Community Survey data. We analyze 12,227 visitations from 1121 block groups to 30 neighborhood parks from September 2019 to August 2020 in Austin, Texas. Time-fixed-effect nested regression models are employed to control for unobserved time-related effects. Findings indicate a substantial increase in overall park use during the pandemic, with visiting frequencies up by 22 % and dwell time by 56 %. In contrast, minority groups, including seniors, children, blacks, unemployed, living alone, and recent movers, exhibited a decline in park visits. It also highlights visitors’ risk-averse behaviors during the pandemic, such as avoiding visiting parks during holiday weeks and avoiding using swimming pools. Before the pandemic, proximity held the utmost significance for park use; during the pandemic, park quality and facilities emerged as the primary factors influencing park utilization. This study suggests that park planning needs to improve neighborhood parks’ proximity, park facilities, and park safety to boost park use and foster inclusive park planning.