Remotely sensed modelling of urban spatial-environmental health across 32 major cities in China

The timely and effective evaluation of spatially-explicit urban PM2.5 concentrations is highly important for an improved understanding of urban environmental health and sustainability in China. However, recent studies examining the spatiotemporal patterns of PM2.5 concentrations has not been comprehensively clarified due to the absence of spatial-detailed urban landscape linkage. In this study, a general non-linear model was developed to depict the non-linear relationship between spatially explicit impervious surface area (ISA) fractions and associated PM2.5 concentrations gradient changes for individual cities from multi-source satellite-based dataset. The comparative results of environmental quality across 32 major cities in China showed that the spatial pattern of urban PM2.5 concentrations is correlated with geographical orientation and socio-economic clusters—high baseline and more balanced states of PM2.5 concentrations are prevalent in North China and the Yangtze delta agglomerations. Temporally, during the period of 2000-2018, most of cities have the path dependency, whereas high concentration of PM2.5 diffused from the original east of 'Hu-Huanyong Line' toward some cities in north-eastern, central and western regions. In addition, our study highlights the optimizing regional economic structure and promoting urban greening construction will be of great significance in sustainable urban environmental health management.