Validation of CAMS-EAC4 and MERRA-2 aerosol reanalysis products for multisatellite atmospheric correction: a case study over Japan with applications to East Asia

Abstract

Accurate atmospheric correction is critical for multi-satellite Earth observation systems, particularly in dryland ecosystems where aerosol variability significantly affects surface reflectance retrievals. Global reanalysis datasets such as CAMS-EAC4 and MERRA-2 provide spatially continuous aerosol parameters, but their accuracy must be validated before operational use. This study validates these products using ground-based SKYNET observations from five varying sites across Japan, which serve as a testbed representing diverse atmospheric conditions, including Asian dust transport events. We compared the Aerosol Optical Depth (AOD) at 550 nm and the Ångström Exponent (AE) by constructing spatiotemporally matched pairs. Preliminary results show that both models reliably reproduce AOD (Pearson’s R > 0.79), making them suitable for total attenuation calculations. However, significant discrepancies and opposing biases were found in AE simulations, which can impact the accuracy of blue/green band surface reflectance retrievals. The validation framework developed in this Japanese case study provides essential error characterization for applying these reanalysis products to multi-satellite atmospheric correction over broader East Asian dryland regions.