Interactions of desert dust and air pollution over the eastern Mediterranean (EM) have been studied, focusing on two distinct dust transport events on 22 and 28 September 2011. The atmospheric chemistry-climate model EMAC has been used at about 50 km grid spacing, applying an online dust emission scheme and calcium as a proxy for dust reactivity. EMAC includes a detailed tropospheric chemistry mechanism, aerosol microphysics and thermodynamics schemes to describe dust "aging". The model is evaluated using ground-based observations for aerosol concentrations and aerosol optical depth (AOD) as well as satellite observations. Simulation results and back trajectory analysis show that the development of synoptic disturbances over the EM can enhance dust transport from the Sahara and Arabian deserts in frontal systems that also carry air pollution to the EM. The frontal systems are associated with precipitation that controls the dust removal. Our results show the importance of chemical aging of dust, which increases particle size, dust deposition and scavenging efficiency during transport, overall reducing the lifetime relative to non-aged dust particles. The relatively long travel periods of Saharan dust result in more sustained aging compared to Arabian dust. Sensitivity simulations indicate 3 times more dust deposition of aged relative to pristine dust, which significantly decreases the dust lifetime and loading.

Abstract. To improve the aeolian dust budget calculations with the global ECHAM/MESSy atmospheric chemistryclimate model (EMAC), which combines the Modular Earth Submodel System (MESSy) with the ECMWF/Hamburg (ECHAM) climate model developed at the Max Planck Institute for Meteorology in Hamburg based on a weather prediction model of the European Centre for Medium-Range Weather Forecasts (ECMWF), we have implemented new input data and updates of the emission scheme.The data set comprises land cover classification, vegetation, clay fraction and topography. It is based on up-todate observations, which are crucial to account for the rapid changes of deserts and semi-arid regions in recent decades. The new Moderate Resolution Imaging Spectroradiometer (MODIS)-based land cover and vegetation data are time dependent, and the effect of long-term trends and variability of the relevant parameters is therefore considered by the emission scheme. All input data have a spatial resolution of at least 0.1 • compared to 1 • in the previous version, equipping the model for high-resolution simulations.We validate the updates by comparing the aerosol optical depth (AOD) at 550 nm wavelength from a 1-year simulation at T106 (about 1.1 • ) resolution with Aerosol Robotic Network (AERONET) and MODIS observations, the 10 µm dust AOD (DAOD) with Infrared Atmospheric Sounding Interferometer (IASI) retrievals, and dust concentration and deposition results with observations from the Aerosol Comparisons between Observations and Models (AeroCom) dust benchmark data set. The update significantly improves agreement with the observations and is therefore recommended to be used in future simulations.

We introduce a framework to efficiently parameterise the aerosol water uptake for mixtures of semi-volatile and non-volatile compounds, based on the coefficient, ν i . This solute-specific coefficient was introduced in Metzger et al. (2012) to accurately parameterise the single solution hygroscopic growth, considering the Kelvin effect -accounting for the water uptake of concentrated nanometer-sized particles up to dilute solutions, i.e. from the compounds relative humidity of deliquescence (RHD) up to supersaturation (Köhler theory). Here we extend the ν i parameterisation from single to mixed solutions. We evaluate our framework at various levels of complexity, by considering the full gas-liquid-solid partitioning for a comprehensive comparison with reference calculations using the E-AIM, EQUI-SOLV II and ISORROPIA II models as well as textbook examples. We apply our parameterisation in the EQuilibrium Simplified Aerosol Model V4 (EQSAM4clim) for climate simulations, implemented in a box model and in the global chemistry-climate model EMAC. Our results show (i) that the ν i approach enables one to analytically solve the entire gas-liquid-solid partitioning and the mixed solution water uptake with sufficient accuracy, (ii) that ammonium sulfate mixtures can be solved with a simple method, e.g. pure ammonium nitrate and mixed ammonium nitrate and (iii) that the aerosol optical depth (AOD) simulations are in close agreement with remote sensing observations for the year 2005. Long-term evaluation of the EMAC results based on EQSAM4clim and ISORROPIA II will be presented separately.

We present a sensitivity study on transatlantic dust transport, a process which has many implications for the atmosphere, the ocean and the climate. We investigate the impact of key processes that control the dust outflow, i.e., the emission flux, convection schemes and the chemical aging of mineral dust, by using the EMAC model following Abdelkader et al. (2015). To characterize the dust outflow over the Atlantic Ocean, we distinguish two geographic zones: (i) dust interactions within the Intertropical Convergence Zone (ITCZ), or the dust–ITCZ interaction zone (DIZ), and (ii) the adjacent dust transport over the Atlantic Ocean (DTA) zone. In the latter zone, the dust loading shows a steep and linear gradient westward over the Atlantic Ocean since particle sedimentation is the dominant removal process, whereas in the DIZ zone aerosol–cloud interactions, wet deposition and scavenging processes determine the extent of the dust outflow. Generally, the EMAC simulated dust compares well with CALIPSO observations; however, our reference model configuration tends to overestimate the dust extinction at a lower elevation and underestimates it at a higher elevation. The aerosol optical depth (AOD) over the Caribbean responds to the dust emission flux only when the emitted dust mass is significantly increased over the source region in Africa by a factor of 10. These findings point to the dominant role of dust removal (especially wet deposition) in transatlantic dust transport. Experiments with different convection schemes have indeed revealed that the transatlantic dust transport is more sensitive to the convection scheme than to the dust emission flux parameterization. To study the impact of dust chemical aging, we focus on a major dust outflow in July 2009. We use the calcium cation as a proxy for the overall chemical reactive dust fraction and consider the uptake of major inorganic acids (i.e., H2SO4, HNO3 and HCl) and their anions, i.e., sulfate (SO42−), bisulfate (HSO4−), nitrate (NO3−) and chloride (Cl−), on the surface of mineral particles. The subsequent neutralization reactions with the calcium cation form various salt compounds that cause the uptake of water vapor from the atmosphere, i.e., through the chemical aging of dust particles leading to an increase of 0.15 in the AOD under subsaturated conditions (July 2009 monthly mean). As a result of the radiative feedback on surface winds, dust emissions increased regionally. On the other hand, the aged dust particles, compared to the non-aged particles, are more efficiently removed by both wet and dry deposition due to the increased hygroscopicity and particle size (mainly due to water uptake). The enhanced removal of aged particles decreases the dust burden and lifetime, which indirectly reduces the dust AOD by 0.05 (monthly mean). Both processes can be significant (major dust outflow, July 2009), but the net effect depends on the region and level of dust chemical aging