Dynamic assessments of population exposure to urban greenspace using multi-source big data.

A growing body of evidence has proven that urban greenspace is beneficial to improve people's physical and mental health. However, knowledge of population exposure to urban greenspace across different spatiotemporal scales remains unclear. Moreover, the majority of existing environmental assessments are unable to quantify how residents enjoy their ambient greenspace during their daily life. To deal with this challenge, we proposed a dynamic method to assess urban greenspace exposure with the integration of mobile-phone locating-request (MPL) data and high-spatial-resolution remote sensing images. This method was further applied to 30 major cities in China by assessing cities' dynamic greenspace exposure levels based on residents' surrounding areas with different buffer scales (0.5km, 1km, and 1.5km). Results showed that regarding residents' 0.5-km surrounding environment, Wenzhou and Hangzhou were found to be with the greenest exposure experience, whereas Zhengzhou and Tangshan were the least ones. The obvious diurnal and daily variations of population exposure to their surrounding greenspace were also identified to be highly correlated with the distribution pattern of urban greenspace and the dynamics of human mobility. Compared with two common measurements of urban greenspace (green coverage rate and green area per capita), the developed method integrated the dynamics of population distribution and geographic locations of urban greenspace into the exposure assessment, thereby presenting a more reasonable way to assess population exposure to urban greenspace. Additionally, this dynamic framework could hold potential utilities in supporting urban planning studies and environmental health studies and advancing our understanding of the magnitude of population exposure to greenspace at different spatiotemporal scales.