AAAR 32nd Annual Conference
September 30 - October 4, 2013
Oregon Convention Center
Portland, Oregon, USA
Abstract View
Understanding and Constraining Global Secondary Organic Aerosol Amount and Size-Resolved Condensational Behavior
STEPHEN D'ANDREA, Silja Häkkinen, Daniel Westervelt, Chongai Kuang, Ezra Levin, Richard Leaitch, Dominick Spracklen, Ilona Riipinen, Jeffrey Pierce, Dalhousie University
Abstract Number: 556 Working Group: Aerosols, Clouds, and Climate
Abstract Secondary organic aerosols (SOA) are major contributors to ultrafine particle growth to climatically relevant sizes in the continental boundary layer (BL). Many models treat SOA solely as semivolatile, which leads to condensation of SOA onto the aerosol mass distribution. However, closure studies with field measurements show that a significant fraction of SOA condenses to the aerosol surface area suggesting that the organics have very low volatilities. Additionally, many global models contain only biogenic sources of SOA (emissions generally 10-30 Tg yr$^(-1)), but recent studies have shown that an additional source of SOA around 100 Tg yr$^(-1) (correlated with anthropogenic CO emissions) may be required to match measurements. Recent studies on sub-20nm particle growth also show the importance of including a size-dependent growth rate parameterization (GRP) to match measurements. We explore the significance of these findings using GEOS-Chem-TOMAS global aerosol microphysics model and observations of aerosol size distributions around the globe. The change in the concentration of particles of size D$_p> 80 nm (N80) within the BL assuming surface area condensation compared to mass distribution condensation yielded a global increase of 9% but exceeded 50% in biogenically active regions. The change in N80 within the BL with the inclusion of the additional SOA compared to the base simulation (19 Tg yr$^(-1)) both using surface area condensation yielded a global increase of 30%, but exceeded 100% in regions with large CO emissions. The inclusion of two different GRP in the additional SOA case both yielded a global increase in N80 of <1%, however exceeded 5% in some cases. These model simulations were compared to measured data obtained from diverse locations around the globe and the results confirmed a decrease in the model-measurement bias when surface-area condensation and the extra SOA were used.