Modeling of carbonaceous aerosols for air pollution health impact studies in Europe

Abstract Air pollution from fine particulate matter (PM2.5) has been associated with various health implications that can lead to increased morbidity and excess mortality. Epidemiological and toxicological studies have shown that carbonaceous particles (black carbon and organic aerosols) may be more hazardous to human health than inorganic ones. Health impact studies and emission reduction policies are based on total PM2.5 concentration without differentiating the more harmful components. In such assessments, PM2.5 and their sub-component concentrations are usually modeled with air quality models. Organic aerosols have been shown to be consistently underestimated, which may affect excess mortality estimates. Here, we use the WRF-Chem model to simulate PM2.5 (including carbonaceous particles) over the wider European domain and assess some of the main factors that contribute to uncertainty. In particular, we explore the impact of anthropogenic emissions and meteorological modeling on carbonaceous aerosol concentrations. We further assess their effects on excess mortality estimates by using the Global Exposure Mortality Model (GEMM). We find that meteorological grid nudging is essential for accurately representing both PM2.5 and carbonaceous aerosols and that, for this application, results improve more significantly compared to spectral nudging. Our results indicate that the explicit account of organic precursors (semi-volatile and intermediate-volatile organic carbons—SVOCs/IVOCs) in emission inventories would improve the accuracy of organic aerosols modeling. We conclude that uncertainties related to PM2.5 modeling in Europe lead to a ∼15% deviation in excess mortality, which is comparable to the risk model uncertainty. This estimate is relevant when all PM2.5 sub-components are assumed to be equally toxic but can be higher by considering their specific toxicity.