Validation of Aerosol Products From Esa/Aatsr Over China and Aod Fusion Based on Uncertainties

Abstract

Abstract. Aerosols play an important role in climate changes and environmental changes as well as on human health. ADV v3.11, ORAC v4.10, and SU v4.32 was three new versions of Advanced Along-Track Scanning Radiometer (AATSR) aerosol datasets which are published on Climate Change Initiative (CCI). In order to evaluate the accuracy of three AATSR aerosol optical depth (AOD) datasets, and to improve the spatial coverage and accuracy of AATSR AOD dataset, this study completed two works: the first part is validated the accuracy of three AOD datasets; the second part is fused three AATSR AOD datasets based on uncertainty of each dataset. After comparing with AERONET and CARSNET ground-based data over China, results show that the root-mean-square error (RMSE) for ADV v3.11, ORAC v4.10, SU v4.32 and fused AOD are 0.22, 0.17, 0.18, 0.17 respectively.

Supplementary material to "LGHAP: a Long-term Gap-free High-resolution Air Pollutants concentration dataset derived via tensor flow based multimodal data fusion"

“…The Advanced Along-Track Scanning Radiometer (AATSR) is one of the Announcement of Opportunity (AO) instruments onboard the European Space Agency (ESA) satellite Envisat, which was primarily designed for measuring sea surface temperature (Wen et al, 2019). The AATSR was designed to have seven spectral channels at wavelengths of 0.55, 0.67, 0.87, 1.63, 10.7, and 12 µm, which observes at dual view as one is the nadir direction and the other is forward direction with a viewing angle of 55 o from nadir view.…”

“…Previous studies had indicated that AOD from AATSR had a generally good accuracy, with R of 0.88 and RMS of 0.24 when compared with AOD observations from AERONET (de Leeuw et al, 2018). In this study, we used ensemble AOD product provided by Climate Change Initiative (CCI) that aggregating three new versions of AATSR aerosol datasets with spatial resolution of 0.1° (ADV v3.11, ORAC v4.10, and SU v4.32) (de Leeuw et al, 2018;Wen et al, 2019;Xie et al, 2018).…”