Comparison of Satellite-based PM2.5 Estimation from Aerosol Optical Depth and Top-of-atmosphere Reflectance

• Published in: Aerosol and air quality research Vol. 21; no. 2; pp. 200257 - 17
• Main Authors: Bai, Heming, Zheng, Zhi, Zhang, Yuanpeng, Huang, He, Wang, Li
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.02.2021; Taiwan Association of Aerosol Research; Springer
• Subjects: Aerosols; Algorithms; Atmosphere; Estimates; Estimation; Machine learning; Meteorological data; Optical analysis; Optical thickness; Original Research; Particulate matter; PM2.5; Population; Probability density functions; Reflectance; Robustness; Root-mean-square errors; Satellite observation; Satellite remote sensing; Spatial distribution; Statistical analysis; Support vector machines; Synchronous satellites; TOA reflectance; Yangtze River Delta; China; Satellite remote sensing; TOA reflectance; PM; Machine learning
• ISSN: 1680-8584; 2071-1409
• DOI: 10.4209/aaqr.2020.05.0257

Aerosol optical depth (AOD) and top-of-atmosphere (TOA) reflectance are two useful sources of satellite data for estimating surface PM 2.5 concentrations. Comparison of PM 2.5 estimates between these two approaches remains to be explored. In this study, satellite observations of TOA reflectance and AOD from the Advanced Himawari Imager (AHI) onboard the Himawari-8 geostationary satellite in 2016 over Yangtze River Delta (YRD) and meteorological data are used to estimate hourly PM 2.5 based on four different machine learning algorithms (i.e., random forest, extreme gradient boosting, gradient boosting regression, and support vector regression). For both reflectance-based and AOD-based approaches, our cross validated results show that random forest algorithm achieves the best performance, with a coefficient of determination (R 2 ) of 0.75 and root-mean-square error (RMSE) of 18.71 µg m −3 for the former and R 2 = 0.65 and RMSE = 15.69 µg m −3 for the later. Additionally, we find a large discrepancy in PM 2.5 estimates between reflectance-based and AOD-based approaches in terms of annual mean and their spatial distribution, which is mainly due to the sampling difference, especially over northern YRD in winter. Overall, reflectance-based approach can provide robust PM 2.5 estimates for both annual mean values and probability density function of hourly PM 2.5 . Our results further show that almost all population lives in non-attainment areas in YRD using annual mean PM 2.5 from reflectance-based approach. This study suggests that reflectance-based approach is a valuable way for providing robust PM 2.5 estimates and further for constraining health impact assessments.