Spatiotemporal dynamics and exposure analysis of daily PM2.5 using a remote sensing-based machine learning model and multi-time meteorological parameters

Identifying spatiotemporal characteristics of daily fine particulate matter (PM2.5) concentrations is essential for assessing air quality and understanding the environmental consequences of urbanization. More importantly, exposure analysis can provide basic information for appropriate decision making. This paper demonstrates an integrated method incorporating satellite-based aerosol optical depth, machine learning models and multi-time meteorological parameters to analyze spatiotemporal dynamics and exposure levels of daily PM2.5 in the economically developed Yangtze River Delta (YRD) from 2016 to 2018. Ten-fold cross validation (CV) was implemented to evaluate the model performance. Compared to the models with daily means of meteorological fields, the models with multi-time meteorological parameters had higher CV coefficient of determination (R-2) and lower CV root mean square error (RMSE) values. The model with the best performance achieved sample-(site-) based CV R-2 values of 0.88 (0.88) and RMSE values of 10.33 (10.35) mu g/m(3). The YRD region was seriously polluted (exceeding the World Health Organization Interim Targets-1 standard of 35 mu g/m(3)) during our study period, especially in Jiangsu Province, but with an improving trend. The residents in Zhejiang Province suffered the least from exposure, with 39 days (4% of the total days) characterized as over polluted (daily average > 75 mu g/m(3)) in our study period. Air pollution in Shanghai Municipality mitigated the most from 2016 to 2018. With the advantages of high-accuracy and high-resolution (daily and 0.01 degrees x 0.01 degrees resolutions), the proposed method can guide for environmental policy planning.