Measuring daytime and nighttime variations in leisure service efficiency of community green spaces for older adults

Abstract

Community green spaces (CGSs) offer a range of leisure services (LSs) for older adults. However, the relationship between LS resources and their effectiveness, particularly regarding efficiency, across day and night scenarios and different geographical contexts remains underexplored. Using Guangzhou, China, as a case study, this research integrates human-perspective CGS imagery, crowdsourced data, and deep learning to quantify LS resources, effectiveness, and efficiency in both daytime and nighttime contexts. The results show that areas with abundant resources or high effectiveness do not necessarily exhibit high efficiency. High-efficiency areas during the day are primarily community parks and high-quality residential green spaces, but shift at night to roadside green spaces and residential green spaces in older urban areas. Conversely, roadside green spaces and residential green spaces within urban villages show low efficiency during the day, whereas at night, community parks and high-quality residential areas exhibit lower efficiency. The study recommends increasing daytime LS resources in urban villages and older neighborhoods and enhancing nighttime resources in community parks and high-quality residential areas. In addition, the findings highlight the nonlinear impact of LS resources on efficiency, with brightness identified as the most influential factor. This study proposes a theoretical model for assessing LS efficiency for older adults and provides practical recommendations for optimizing CGS resource allocation to support creation of age-friendly environments.