1Norwegian Institute for Air Research (NILU), P.O. Box 100, 2027 Kjeller, Norway
2Norwegian Meteorological Institute, Oslo, Norway
3Umweltbundesamt, Langen, Germany
4University of Gothenburg, Gothenburg, Sweden
5Swedish Meteorological and Hydrological Institute, Norrköping, Sweden
6School of Physics & Center for Climate and Air Pollution Studies, Ryan Institute, National University of Ireland Galway, Galway, Ireland
7National Environmental Research Institute (NERI), Roskilde, Denmark
8Umweltbundesamt, Vienna, Austria
9Air Pollution, Environmental Technology EMPA, Dübendorf, Switzerland
10Finnish Meteorological Institute (FMI), Helsinki, Finland
11Centre for Ecology and Hydrology (CEH), Edinburgh, UK
12Energy Centre of the Netherlands (ECN), Petten, The Netherlands
13Institute of Environmental Assessment and Water Research (IDAEA-CSIC), Barcelona, Spain
14CNR, The Institute for Atmospheric Pollution, Rome, Italy
15Paul Scherrer Institut (PSI), Villigen, Switzerland
16European Commission – DG Joint Research Centre, Ispra, Italy
17Chalmers University of Technology, Gothenburg, Sweden
18Leibniz Institute for Tropospheric Research (IfT), Leipzig, Germany
19The Czech Hydrometeorological Institute (CHMI), Prague, Czech Republic
Abstract. The first EMEP intensive measurement periods were held in June 2006 and January 2007. The measurements aimed to characterize the aerosol chemical compositions, including the gas/aerosol partitioning of inorganic compounds. The measurement program during these periods included daily or hourly measurements of the secondary inorganic components, with additional measurements of elemental- and organic carbon (EC and OC) and mineral dust in PM1, PM2.5 and PM10. These measurements have provided extended knowledge regarding the composition of particulate matter and the temporal and spatial variability of PM, as well as an extended database for the assessment of chemical transport models. This paper summarise the first experiences of making use of measurements from the first EMEP intensive measurement periods along with EMEP model results from the updated model version to characterise aerosol composition. We investigated how the PM chemical composition varies between the summer and the winter month and geographically.
The observation and model data are in general agreement regarding the main features of PM10 and PM2.5 composition and the relative contribution of different components, though the EMEP model tends to slightly underestimate PM10 and PM2.5 compared to measurements. The intensive measurement data has identified areas where improvements are needed. In particular, the model description of formation of coarse nitrate on sea salt and dust particles requires further attention. Hourly concurrent measurements of gaseous and particulate components for the first time facilitated testing of modelled diurnal variability of the gas/aerosol partitioning of nitrogen species. In general, the modelled diurnal cycles of nitrate and ammonium aerosols are in good agreement with the measurements. As expected, the diurnal variability of ammonia is not very well captured, but this will probably improve if the EMEP model is coupled to a dynamic, mechanistic ammonia emission module. The largest underestimations of aerosol mass are seen in Italy during winter, which to a large extent may be explained by an underestimation of residential wood burning source. It should be noted that both primary and secondary OC has been included in the calculations for the first time, showing promising results. Mineral dust is important, especially in southern Europe, and the model seems to capture the dust episodes well. The lack of measurements of mineral dust hampers the possibility for model evaluation for this highly uncertain PM component.
There are also lessons learnt regarding improved measurements for future intensive periods. There is a need for increased comparability between the measurements at different sites. For the nitrogen compounds it is clear that more measurements using artefact free methods based on continuous measurement methods and/or denuders are needed. For EC/OC, a reference methodology (both in field and laboratory) was lacking during these periods giving problems with comparability, though presently measurement protocols have been established and these should be followed by the Parties to the EMEP Protocol. For measurements with no defined protocols, it might be a good solution to use centralised laboratories to ensure comparability across the network. To cope with the introduction of these new measurements new reporting guidelines have been developed to ensure that all proper information about the methodologies and data quality is given.