In recent years, implementation of aggressive and strict clean air policies has resulted in significant decline in observed PM2.5 concentration in the Beijing-Tianjin-Hebei (BTH) region and its surrounding areas (i.e., the "2 + 26" region). To eliminate the effects of interannual and seasonal meteorological variation, and to evaluate the effectiveness of emission abatement policies, we applied a boosted regression tree model to remove confounding meteorological factors. Results showed that the annual average PM2.5 concentration normalized by meteorology for the "2 + 26" region declined by 38% during 2014-2019 (i.e., from 96 to 60 μg/m3); however, the BTH region exhibited the most remarkable decrease in PM2.5 concentration (i.e., a 60% reduction). Certain seasonal trend in normalized PM2.5 level remained for four target subregions owing to the effects of anthropogenic emissions in autumn and winter. Although strong interannual variations of meteorological conditions were unfavorable for pollutant dispersion during the heating seasons of 2016-2018, the aggressive abatement policies were estimated to have contributed to reductions in normalized PM2.5 concentration of 19%, 10%, 19%, and 17% in the BTH, Henan, Shandong, and Shanxi subregions, respectively. Our study eliminated the meteorological contribution to concentration variation and confirmed the effectiveness of the implemented clean air policies.
Keywords: Boosted regression tree model; Emission abatement; Meteorological confounding; PM(2.5).
Copyright © 2020 Elsevier Ltd. All rights reserved.