重庆市绿色空间景观格局与 PM2.5 浓度时空相关性.

Airborne fine particulate matter poses a serious threat to human health, and investigating the impact of green space landscape patterns on PM2.5 concentrations is conducive to reducing the risk of respiratory diseases by adjusting the green space pattern to decrease PM2.5 concentrations. Utilizing land use remote sensing monitoring data and PM2.5 concentration data in Chongqing city from 1980 to 2020 as the foundational dataset, the landscape pattern index method and spatial autocorrelation analysis were employed to study the characteristics of green space landscape patterns and PM2.5 concentration changes. Furthermore, a geographical and temporal weighted regression(GTWR)model was applied to explore the influence of changes in green space landscape pattern indices on PM2.5 concentrations and their spatiotemporal heterogeneity. The results show that PM2.5 concentrations in Chongqing city gradually increase from 1980 to 2010, followed by a gradual decrease from 2010 to the present; simultaneously, the spatial distribution exhibits significant agglomeration characteristics, primarily manifesting as low-low agglomeration in the east and high-high agglomeration in the west. Area index(TA), patch density index(PD), and patch cohesion index(COHESION)of forest, grassland and agricultural land show significant correlations with PM2.5 concentrations; specifically, area index of forest is negatively correlated, while area indexes of arable land and grassland are positively correlated; patch density indexes of forest and grassland are positively correlated, while patch density index of arable land is negatively correlated; patch cohesion indexes of forest, grassland and arable land are all negatively correlated. In the metropolitan area, area index of grassland and patch density index of arable land exhibit a stronger negative impact on PM2.5 concentrations; in the urban clusters of Three Gorges reservoir area in the northeast and Wuling mountain area in the southeast of Chongqing city, aggregation index(AI), patch density index, patch cohesion index of forest, and area index of arable land exert a stronger influence on PM2.5 concentrations. [ABSTRACT FROM AUTHOR]