Numerical simulation and evaluation of global ultrafine particle concentrations

Ultrafine particles (UFP), i.e. particles with an aerodynamic diameter below 100 nm, have a significant impact on public health and the hydrological cycle. Recent studies showed that their ability to penetrate more deeply into the lungs and potentially into the blood stream may cause an increased incidence of cardiovascular and cerebrovascular diseases. Additionally, UFPs significantly contribute to cloud condensation nuclei concentrations. However, knowledge on global distributions of UFPs is scarce.

We present a global simulation of UFP concentrations using the ECHAM/MESSy Atmospheric Chemistry Model (EMAC), including tropospheric and middle-atmospheric processes, and the modal aerosol microphysics submodel GMXe. Due to the high sensitivity of UFP concentrations to the size distribution of emitted particles, we derived emission median diameter for primary emissions from various sectors and species based on existing literature. We show the importance of primary emissions and nucleation on UFP concentrations as well as their composition, seasonality and vertical distributions.

Model results were evaluated over Europe, the United States, India and China, using particle size distribution and particle number concentration measurements from available datasets and the literature. We obtain reasonable agreement between the model results and observations. However, the highest values of observed, street-level UFP concentrations are systematically underestimated, whereas in rural environments close to urban areas they are generally overestimated by the model. As the relatively coarse global model does not resolve concentration gradients in urban centres and local UFP hotspots, high-resolution data of anthropogenic emissions is used to account for such differences in each model grid box. This downscaling further improves the agreement with observations, decreasing the root mean squared logarithmic error and removing discrepancies associated with air quality and population density gradients within the model grid boxes.