Cerca nel sito per parola chiave

pubblicazioni - Articolo

Modelling Reconstruction of PM Composition in the Po Valley

pubblicazioni - Articolo

Modelling Reconstruction of PM Composition in the Po Valley

Guido Pirovano*, Giuseppe Maurizio Riva* ACCENT/GLOREAM – 22nd Workshop on tropospheric chemical transport Brescia, 26-27 Novembre 2009 PRESENTAZIONE POWER POINT * ERSE SpA The Po valley, a flat area with peculiar orographical and meteorological features placed in Northern Italy, often suffers high PM concentrations, whose modeling reconstruction is a challenging task, usually putting in evidence a clear underestimation of the PM10 bulk concentration. Due to a lack in operational measurements of particulate matter compounds, modeling studies generally fail in explaining the reason of such discrepancies. In order to provide further investigations of the PM phenomena, the CAMx chemistry transport model has been applied in the framework of the POMI project (http://aqm.jrc.it/POMI/), where a comprehensive meteorological and chemical measured dataset was available. The CAMx model was used for simulation over the Po valley air basin with a 6 x 6 km 2 spatial resolution for the whole 2005. Meteorological fields have been compared to a set of 115 stations belonging to both regional networks and also the national network of WMO Synop stations. Model results for gas species and PM10 bulk concentration have been compared to a set of 64 representative stations, belonging to the regional operational networks. Moreover speciated PM2.5 mass concentration have been compared to a field campaign performed over the Lombardy region. Model performance statistics highlighted that, on average, CAMx correctly reproduce both NO X and O 3 concentrations. This positive result is partially related to error compensation in the meteorological fields, where wind speed is slightly overestimated, while vertical diffusion is probably underestimated in urban areas. Temporal evolution of daily PM10 concentration is well reproduced by the model, whereas the yearly mean bias is around 50%. NO X and PM10 statistics at each station are well correlated, but PM10 performance are systematically worse than NO X , confirming that discrepancies with observed PM10 are not only related to dispersion. Daily time series put in evidence that PM10 underestimation takes place both during winter and summer. Summer underestimation could be related to a lack in the reconstruction of humidity that might be unfavorable to the formation of secondary PM. Differently; winter underestimation is probably due to some missing sources in the PM emission inventory data. Speciated PM results point out that PM mass underestimations essentially derive from a lack of organic matter, whereas a rather good agreement is observed for the other chemical species and especially for the ammonium-nitrate system. Such results suggest to put particular attention to sources like domestic heating wood burning, and to consider fine PM formation mechanism form condensable gaseous precursors, directly emitted in the atmosphere, whose contribution may account for large part of the missing PM mass on the cold season.

Progetti

Commenti