Effects of chemical aging on global secondary organic aerosol using the volatility basis set approach

Abstract

A global 3-D chemical transport model (GEOS-Chem) is used with the volatility basis set (VBS) approach to examine the effects of chemical aging on global secondary organic aerosol (SOA) concentrations and budgets. We present full-year simulations and their comparisons with the global aerosol mass spectrometer (AMS) dataset, the Interagency Monitoring of Protected Visual Environments (IMPROVE) dataset from the United States, the European Monitoring and Evaluation Programme (EMEP) dataset from Europe, and water-soluble organic carbon observation data collected over East Asia. Using different chemical aging constants, we find that the model results with 4 × 10−11 cm3 molecule−1 s−1 are in better agreement with all observations relative to the model results with other aging constants, without aging, and with the two-product approach. The model simulations are improved when chemical aging is considered, especially for rural regions. However, the simulations still underestimate observed oxygenated organic aerosol (OOA) in urban areas. Two sensitivity simulations including semi-volatile primary organic aerosol (POA) were conducted. We find that including semi-volatile POA improves the model in terms of the hydrogen-like organic aerosol (HOA) to OOA ratio. However, the total OA concentrations are not improved. The total SOA production is considerably increased by 53%, from 26.0 to 39.9 Tg yr−1, after considering chemical aging, remaining lower than top-down estimates of SOA production. Direct radiative forcing (DRF) increases by −0.07 W m−2 due to the chemical aging of SOA, which is comparable to the mean DRF (−0.13 W m−2) of OA from the AeroCom multi-model study. This result indicates considerable global and, more importantly, regional climate implications. For example, the regional DRF change due to chemical aging of SOA in the eastern US is −0.29 W m−2, which is 4 times greater in magnitude than the global mean value.