Uncertainties in global aerosol simulations: Assessment using three meteorological data sets

[1] Current global aerosol models use different physical and chemical schemes and parameters, different meteorological fields, and often different emission sources. Since the physical and chemical parameterization schemes are often tuned to obtain results that are consistent with observations, it is difficult to assess the true uncertainty due to meteorology alone. Under the framework of the NASA global modeling initiative (GMI), the differences and uncertainties in aerosol simulations (for sulfate, organic carbon, black carbon, dust, and sea salt) solely due to different meteorological fields are analyzed and quantified. Three meteorological data sets available from the NASA Goddard Data Assimilation Office (DAO) general circulation model (GCM), the Goddard Institute for Space Studies (GISS) GCM, version II' and the NASA Goddard Global Modeling and Assimilation Office (GMAO), finite-volume GCM (FVGCM) are used to drive the same aerosol model. The global sulfate and mineral dust burdens with FVGCM fields are 40% and 20% less than those with DAO and GISS fields, respectively, due to its larger precipitation. Meanwhile, the sea salt burden predicted with FVGCM fields is 56% and 43% higher than those with DAO and GISS, respectively, due to its stronger convection especially over the Southern Hemispheric Ocean. Sulfate concentrations at the surface in the Northern Hemisphere extratropics and in the middle to upper troposphere differ by a factor of 3 between the three meteorological data sets. The agreement between model calculated and observed aerosol concentrations in the surface source regions is similar for all three meteorological data sets. Away from the source regions, however, the comparisons with observations differ greatly for DAO, FVGCM, and GISS, and the performance of the model using different meteorological data sets varies depending on the site and the compared species. Sensitivity simulations with the NASA GEOS-4 assimilated fields show that the interannual variability of aerosol concentrations can be higher than a factor of 2 depending on the location and season, which is generally, however, smaller than the differences due to using different meteorological data sets. Global annual average aerosol optical depth at 550 nm is 0.120–0.131 for the three meteorological data sets. However, the contributions from different aerosol components to this total optical depth differ significantly, which reflects differences in the aerosol spatial distributions. The global annual average anthropogenic and all-sky aerosol direct forcing at the top-of-the atmosphere is estimated to be −0.75, −0.35, and −0.40 W m−2 for DAO, FVGCM, and GISS fields, respectively. Regional differences can be much larger (by a factor of 4–5) in the tropics over the ocean and in the polar regions.