NO2, PM10 and O3 urban concentrations and its association with circulation weather types in Portugal

High levels of atmospheric pollutants are frequently measured in Portugal, a country which has been affected by several pollution episodes, exceeding PM10, O3 and NO2 legal limits repeatedly during the last decade. The occurrence of these episodes is often related to either local-scale conditions or regional-scale transport. In order to better understand the atmospheric factors responsible for poor air quality, the relationships between air pollution and meteorological variables or atmospheric synoptic patterns represent an important research area. Here an objective classification scheme of the atmospheric circulation affecting Portugal, between 2002 and 2010, is presented, where daily circulation is characterized through the use of a set of indices associated with the direction and vorticity of the geostrophic flow in the lower atmosphere. The synoptic characteristics and the frequency of ten basic circulation weather types (CWTs) are discussed and a framework that permits the identification of the main characteristics associated to the occurrence of pollution episodes is mapped based on the identified patterns. The relationship between CWTs and poor air quality allowed distinguishing between which types are most frequently associated to pollution episodes. It is shown that the anticyclonic and north types, although being the most frequent classes during the majority of the year, do not prevail during pollution episodes that are dominated by easterly types. In general, higher concentration of all three pollutants and the two extreme events analysed occur associated predominantly with synoptic circulation characterized by an eastern component and advection of dry air masses. Moreover, results on the link between CWTs and air quality for Lisbon and Porto urban areas suggest that air quality regimes are generally similar for the northern and southern regions considered with the exception of spring and autumn PM10. Results obtained highlight the existence of strong links between the interannual variability of daily air quality and interannual variability of CWTs