Role of Organic Aerosol Chemistry Schemes on Particulate Matter Modeling in Europe

Abstract

Organic aerosols (OA) contribute to 20–90% of the particulate matter (PM) concentration worldwide. To investigate the effects of OA schemes on PM modeling, we conducted whole-year (2010) simulations in Europe using the regional air quality model CAMx v6.50 with three different OA schemes: (1) an updated version of the traditional Odum 2-product model SOAP2.1, (2) the standard volatility basis set (VBS) algorithm, and (3) a modified VBS scheme which applies 3 times of POA emissions to offset the influence of missing semi-volatile organic compound (SVOC) emissions and enables aging of secondary condensable gases from biomass burning (referred as VBSNEW). The modeled OA concentration by SOAP was significantly higher than OA by VBS over land especially in the eastern Europe, while VBS led to higher OA only over the Mediterranean sea. The higher OA–SOAP mainly came from the primary organic aerosol (POA), which was 0.4 µg m−3 higher than POA–VBS on average. However, the SOA–VBS was higher than SOA–SOAP in most of the areas except for the northeastern Europe, where the modeled OA concentration was strongly influenced by the boundary conditions. The VBSNEW increased the domain average OA by 26% and 63% compared to SOAP and VBS, respectively. Both POA and SOA were increased by VBSNEW, especially in winter when biomass burning is the dominant OA source. Comparison with observed PM2.5 in 432 AIRBASE stations in Europe showed that VBSNEW led to lowest mean bias (MB) and root-mean-square error (RMSE), which reduced MB by 1.3 and 0.7 μg m−3 compared to VBS and SOAP.