PM2.5

spatiotemporal distribution

spatial autocorrelation

geographic detector

driving factors

Atmospherical pollution control

Pollution prevention

Environmental management

Policies for environmentally sustainable development

The impact of PM2.5 pollution on the ecological environment and human health and safety has attracted worldwide attention. Previous analyses of large-scale PM2.5 spatiotemporal distribution have mainly been based on satellite observations. Due to the limitations of inversion methods, the real temporal and spatial variation of PM2.5 concentrations cannot be obtained from satellite data. Using PM2.5 data from 2014 to 2017 inversed from the daily PM2.5 data in Shandong Province and employing time series, spatial autocorrelation, and geographical detector methods, the temporal and spatial evolution of PM2.5 concentrations in Shandong Province and their driving factors are revealed. The results show that the annual variation trend in the PM2.5 concentration in Shandong Province is downward, with an average annual concentration of 57.6 μg/m³, while the monthly and winter months show significant high-level u-type changes, with small changes in spring and summer, and the number of days for air compliance in 2017 was 294 . Linyi and Dezhou have the largest annual decline in air quality, while coastal cities have a small decline in air quality. From 2014 to 2017, PM2.5 of Shandong Province showed obvious spatial agglomeration and disparity patterns. The correlation between PM2.5 and other elements differs significantly by month, and soot emission has the most significant effect on PM2.5 concentration. Geographic detection analysis indicates that the main driving factors for the change in PM2.5 concentration in Shandong Province are crop broadcast area and soot emissions.