Resumen:
In late March to early April 2024, an unusually high amount of sand dust was wind-blown to Europe from the Sahara Desert. Most of mainland Europe was affected by these sand dust particles. As a result, Central Europe experienced an exceptionally high increase in air pollution. In this work, the impact of this Saharan dust event on PM10 characteristics in an urban and a natural locality in the Czech Republic was investigated. PM10 concentrations during the Saharan dust event were about 6-8 times higher than under normal atmospheric conditions, exceeding WHO guidelines by up to 2 times. Terrain and altitude may have influenced the local concentrations of Saharan dust. Airborne dust collected before and during the Saharan dust event was then studied using scanning electron microscopy combined with energy-dispersive spectroscopy. These methods were employed to determine the sizes and elemental compositions of the individual dust particles. Further, X-ray diffraction analysis was carried out to reveal the mineralogical composition of the collected dust. Surprisingly, the particle size distribution was not significantly affected by the windblown Saharan dust, but its dependency on the sampling locality was revealed. It may be explained by the different altitudes of the sampling localities, as coarse particles are more susceptible to gravitational pull while fine particles tend to remain suspended at higher altitudes. The dominant mineral in the Saharan dust was calcite, which substantially altered the local PM10 composition. The studied Saharan dust originated from a natural area, as the amount of anthropogenic pollutants detected was negligible. Notably, its carbon content was lower compared with the usual local PM10. The elevated PM10 concentrations appear to be the most relevant risk associated with this Saharan dust event in Central Europe. The transported dust originated from the northern/north-western Sahara - probably from the Atlas region - which was verified by a backward trajectory analysis of air masses using the HYSPLIT model.
