Multiscale analysis of the effects of urban green infrastructure landscape patterns on PM2.5 concentrations in an area of rapid urbanization

Urban green infrastructure (UGI) is considered to be an effective tool for mitigating PM2.5 (particulate matter with an aerodynamic diameter of less than 2.5 μm) pollution in urban areas. However, long and continuous time series analyses of the relationships between the UGI landscape and PM2.5 pollution remain a challenge. In the present study, then we analyzed the PM2.5 variations and their relationships with the UGI landscape patterns at the urban agglomeration (the urban area of CLUA) and neighborhood scales (6 neighborhood zones with radii ranging from 500 m to 3000 m around the sampling point.). The results illustrated PM2.5 concentration increased slightly in the urban expansion areas and decreased in some old urban areas, but the changing trend was not significant. The PM2.5 concentrations were different among the four seasons in this area, and the lowest and highest concentrations occurred in winter (66.79 μg/m³) and summer (24.32 μg/m³), respectively. At the urban agglomeration scale, just a very small proportion of the PM2.5 variations in the CLUA during this study period could be attributed to the UGI landscape pattern, wind speed and relative humidity; PM2.5 concentrations were affected more strongly by wind speed and relative humidity than by UGI landscape patterns; the LSI (landscape shape index) and bridges served as the main landscape indicators affecting the PM2.5 concentrations among the UGI landscape patterns factors. At the neighborhood scale, the LPI (largest patch index), AWMSI (area-weighted mean shape index), AI (aggregation index), Core and Loop served as the main UGI landscape indicators influencing PM2.5 and had different degrees of influence on the PM2.5 concentrations; the seasonal and scale effects of UGI landscape patterns on PM2.5 concentrations at neighborhood scales were observed. Comparatively, the effects of the UGI landscape patterns on PM2.5 concentration seem more significant at the neighborhood scale than at the urban agglomeration scale. The findings of this multiscale analysis can provide new insights into understanding the relationships between UGI landscape patterns and urban air pollution at different scales, and provide a scientific reference for urban planning and air pollution control.