Multi-scale spatiotemporal trends and corresponding disparities of PM_2.5 exposure in China

Despite the effectiveness of targeted measures to mitigate air pollution, China-a developing country with high PM_2.5 concentration and dense population, faces a high risk of PM_2.5-related mortality. However, existing studies on long-term PM_2.5 exposure in China have not reached a consensus as to which year it peaked during the "initially pollution, then mitigation" process. Furthermore, analyses in these studies were rarely undertaken from multi-spatial scales. In this study, a piecewise linear regression model was employed to detect the turning point of population-weighted exposure (PWE) to PM_2.5 for the period 2000-2020. Multi-scale spatiotemporal patterns of PM_2.5 exposure were evaluated during upward and downward periods at the province, city and county levels, and their corresponding disparities were estimated using the Gini index. The results showed that 2013 was the breakpoint year for PM_2.5 PWE across China from 2000 to 2020. Cities and counties where PM_2.5 PWE displayed increasing trends during the mitigation stage (2013-2020) basically became the heaviest PM_2.5 exposure regions in 2020. High PM_2.5 exposure was observed in Beijing-Tianjin-Hebei, Central China, and the Tarim Basin in Xinjiang, whereas lower PM_2.5 exposure regions were mainly concentrated in Hainan Province, the Hengduan Mountains, and northern Xinjiang. These cross-provincial patterns might have been overlooked when conducting macro-scale analyses. Province-level PM_2.5 exposure inequality was less than the city- and county-levels estimations, and regional inequalities were high in eastern and western China. In this study, multi-scale PM_2.5 exposure trends and their disparities over a prolonged period were investigated, and the findings provide a reference for pollution mitigation and regional inequality reduction.