Development of a secondary organic aerosol formation mechanism: comparison with smog chamber experiments and atmospheric measurements

A new mechanism to simulate the formation of secondary organic aerosols (SOA) from reactive primary hydrocarbons is presented, together with comparisons with experimental smog chamber results and ambient measurements found in the literature. The SOA formation mechanism is based on an approach using calculated vapor pressures and a selection of species that can partition to the aerosol phase from a gas phase photochemical mechanism. The mechanism has been validated against smog chamber measurements using ?-pinene, xylene and toluene as SOA precursors, and has an average error of 17%. Qualitative comparisons with smog chamber measurements using isoprene were also performed. A comparison against SOA production in the TORCH 2003 experiment (atmospheric measurements) had an average error of only 12%. This contrasts with previous efforts, in which it was necessary to increase partition coefficients by a factor of 500 in order to match the observed values. Calculations for rural and urban-influenced regions in the eastern U.S. suggest that most of the SOA is biogenic in origin, mainly originated from isoprene. A 0-dimensional calculation based on the New England Air Quality Study also showed good agreement with measured SOA, with about 40% of the total SOA from anthropogenic precursors. This mechanism can be implemented in a general circulation model (GCM) to estimate global SOA formation under ambient NOx and HOx levels.