The impact of Sahara dust on air quality and public health in European countries

The frequent transport of Sahara dust toward Europe degrades the air quality and poses risk to human health. In this study we use GEOS-Chem (a global transport model) to examine the impact of Sahara dust on air quality and the consequent health effect in Europe for the year 2016–2017. The simualtion is conducted in a nested model with the native resolution of 0.25° × 0.3125° (Latitude × Logitude) over Europe (32.75°N–61.25°N, 15°W–40°E). The simulation on a global scale with a coarse horizontal resolution of 2° × 2.5° is also conducted to provide the boundary condition for the nested-grid simulation as well as aerosol optical depth (AOD) over the Sahara desert for model evaluation.

The model performance is evaluated by comparisons with surface observations including aerosol optical depth (AOD) from AERONET, and PM_2.5 and PM₁₀ concentrations from numerous air quality monitoring stations in European countries. Overall, the model well reproduces observed surface PM concentrations over most European countries with some underestimation in southern Europe. In addition, model AOD is highly correlated with AERONET data over both Sahara and European region.

The spatial distribution of dust concentrations, frequency of dust episodes, as well as the exposure and health effects are studied. The concentrations of Sahara dust decrease from 5–20 μg m^-3 in south to 0.5–1.0 μg m^-3 in north of Europe. Spain and Italy are most heavily influenced by Sahara dust in terms of both concentration levels and frequencies of occurrence. Strong dust episodes (>50 μg m^-3) occur predominately in Southern Spain and Italy with frequency of 2–5%, while light dust episodes (>1 μg m^-3) are often detected (5–30%) in Central and Western Europe.

The population-weighted dust concentrations are higher in Southern European countries (3.3–7.9 μg m^-3) and lower in Western European countries (0.5–0.6 μg m^-3). The health effects of exposure to dust is evaluated based on population attributable fraction (PAF). We use the relative risk (RR) value of 1.04 (95% confidence intervals: 1.00 – 1.09) per 10 µg m^-3 of dust exposure based on the main model of Beelen et al. (2014). We estimate a total of 41884 (95% CI: 2110–81658) deaths per year attributed to the exposure to dust in the 13 European countries studied. Due to high contribution to PM₁₀ in Spain, Italy and Portugal, dust accounts for 44%, 27% and 22% of the total number of deaths linked to PM₁₀ exposure, respectively.