Estimating PM 1 concentrations from MODIS over Yangtze River Delta of China during 2014–2017

Abstract

© 2018 Elsevier Ltd Compared to the space-borne estimation of PM 2.5 (particulate matter with aerodynamic diameter ≤2.5 μm), the investigation of PM 1 (≤1 μm) remains less intensive and thus unclear. Here we estimated four years (2014–2017) of ground-level PM 1 concentrations from MODIS aerosol optical depth (AOD) in attempt to gain a better understanding of much finer particles. The Yangtze River Delta (YRD) region, with a relatively dense ground-based PM 1 station network, was selected as the study area. The geographically and temporally weighted regression (GTWR) model simultaneously accounting for spatial and temporal variability existing within various predictors was constructed. Validation of satellite-estimated PM 1 against ground-measured PM 1 yields a high consistence, significant improvement over previous work (R 2 = 0.74 VS 0.59, RMSE = 13.02 μg/m 3 VS 22.5 μg/m 3 ). This suggests the PM 1 estimates from GTWR model are reliable and robust enough to obtain large-scale fine particle contents. The population exposure of air pollution in the YRD region, therefore, has been analyzed by calculating population-weighted mean PM 1 concentrations, which reaches as high as 37.22 μg/m 3 . Further analysis indicates that near half the people live in locations with high-level PM 1 concentration (>35 μg/m 3 ), which has profounding implication for improving our understanding of human exposure to fine aerosol particles.