An Evaluation of the GOES-R ABI Aerosol Retrieval Algorithm

A multi-channel aerosol retrieval algorithm has been developed for the Advanced Baseline Imager (ABI) onboard the next generation US Geostationary Operational Environmental Satellite (GOES-R series). Testing and evaluation of this algorithm is carried out with model simulated and satellite-observed multi-spectral reflectance. Sensitivity tests with synthetic reflectance show that ABI retrieved aerosol optical depth at 0.55 mum (tau_{0.55 mum}) is expected to change by 5% and 9% (14% and 24%) on average over water (land) corresponding to 3% and 5% uncertainties in calibration. Model identification is expected to have a 50%-60% success rate for a 3%-5% calibration uncertainty. Performance of ABI aerosol retrieval is evaluated with MODIS observations. Compared with AERONET ground measurements, on average, retrieved tau_{0.55 mum} differs from AERONET measurements by -0.02plusmn0.05 over water and 0.05plusmn0.17 over land. The effect of sensor polarization sensitivity on ABI aerosol retrievals is also estimated as a bias of -0.01 over water and -0.05 over land for the retrieved tau_{0.55 mum}.