Contribution Of Typical Atmospheric Emission Sources To The Formation Of Organic Aerosol In Pearl River Delta Region

In recent years, the rapid economic development and urbanization have already ledsevere air quality deterioration to the Pearl River Delta （PRD） regions. The PRD city clusterare appearing typical regional and complex air pollution characteristics, with frequent hazedays and photochemical smog pollution. Organic aerosol generally accounts for20-50%ofthe total fine particles in the PRD region, and has been associated with human health,atmospheric visibility, acid precipitation, and climate change. In this study, a newly developedregional emission inventory and the multi-scale air quality simulation model were usedtogether to identify the source contribution of primary organic aerosol and secondary organicaerosol.First, the primary carbonaceous aerosol emissions were estimated based on the collectedactivity data and the latest emission factors for the PRD region in2009. The total PM_2.5, BCand OC emissions were302.5kt,38.7kt and30.9kt, respectively. OC emissions were mainlyfrom residential combustion and mobile sources, with a contribution of34.1%and29.2%,respectively. The main contributor to BC emissions was mobile sources, accounting for about65.0%. The primary OC/BC ratios for individual cities in the PRD region were dependent onthe levels of economic development, with high ratios in the less developed surrounding citiesand low ratios in central and southern developed areas. Guangzhou, Shenzhen and Jiangmenwere the cities of largest carbonaceous emissions. We compared our estimates with othersimilar emission inventories, and did the qualitative and quantitative uncertainty analysis,which indicated that emission inventory developed in our study were reasonable in a way, butstill need to improve.Second, a highly resolved temporal and spatial PRD regional emission inventory wasdeveloped and can be used to support the regional air quality modeling. The preliminarytemporal profiles were established for the important emission sources like power plant,industrial, on-road mobile sources, open biomass burning etc.. The GIS （GeographicInformation System）–based approach was adopted for allocating the PRD regional emissionsby using appropriate spatial surrogates, including geographic coordinate information,population density, road network, inland waterway network, and land use area data etc..Third，both qualitative and quantitative methods were used to evaluate the uncertainty inemission estimates. There were relatively high uncertainties in fine particle and carbonaceousaerosol emission estimates due to lack of emission information. Uncertainty in SO2and NOXwere relatively low because more attentions have been paid on these two pollutants. In addition, uncertainties in VOC and NH3will certainly be underestimated.Finally, three typical emission sources were chosen to set corresponding sensitivity case,including industrial sources, on-road mobile sources, and biogenic VOC. Modeling studieswere conducted to estimate the contribution of these three emission sources to SOA formationwith the aid of air quality modeling system （WRF-SMOKE-CMAQ）. The results showed thatobviously decrease of SOA concentration was observed after shutting down the on-roadmobile sources and biogenic VOC. The contribution of on-road mobile sources to theformation of SOA is higher than other two sources. The industrial sources show highcontributions to PM_2.5than other two emission sources. It can be expected that industrialsources may have important effects on primary particles or secondary inorganic aerosol.