Short-term climatology of PM10 at a high altitude background station in southern Europe

The paper analyzes PM10 data from Mt. Cimone observatory (44 degrees 11' N, 10 degrees 42' E) in the period 1998-2011. Mt. Cimone is the highest peak of the Italian northern Apennines (2165 m as!) which hosts a high altitude background station in Southern Europe. The dataset is discussed in the framework of the main atmospheric meteorological and territorial features, representative of the central Mediterranean free troposphere. The overall geometric mean of PM10 mass concentration in the period investigated is 6.0 mu g m(-3) (arithmetic mean 8.8 +/- 8.0 mu g m(-3)), with large variability at the synoptic time-scale. At Mt. Cimone station the PM10 mass load features a strong seasonal fluctuation with a winter minimum (0.1 mu g m(-3), average winter value: 4.1 +/- 4.8 mu g m(-3)) and a summer maximum (45.7 mu g m(-3), average summer value: 13.7 +/- 7.2 mu g m(-3)). Influence of surface source areas upon PM10 is discussed by comparison of simultaneous PM10 data collected at ground stations to the north and south of the Northern Apennine range in order to investigate transport connections at the regional scale. PM10 data collected from the Environmental Protection Agencies networks of Emilia Romagna and Tuscany mostly cover urban stations and a few rural and semirural stations usually at ground level. In general PM10 mass load exhibits a vertical negative gradient with altitude, with Mt. Cimone displaying the least average value. Nevertheless exceptions can be observed on an event basis, while direct comparison of concentration data is not advisable due to the remoteness of Mt. Cimone as compared even to rural or semirural stations. Analysis of PM10 time series in conjunction with fine and coarse particle number densities reveals the influence of transports from source regions such as the Northern African desert, the Po valley and the European continent including the Balkan area. Source apportionment of PM10 is achieved applying a methodology based on Hysplit back-trajectories calculation. (C) 2012 Elsevier Ltd. All rights reserved.