Air quality and integration of short-term and long-term pollutant data

Antonella Plaia, Anna Lisa Bondi'

Risultato della ricerca: Other


Modelling PM10 is an important problem in statistical methodology, above all to explain the PM10 behaviour in space and time, since it has been linked to many adverse effects on human and environmental health. But the large spatial variability of the main traffic-related pollutants, and in particular here the PM10, implies the impossibility of obtaining from the data of the fixed stations a complete pictures of the atmospheric pollution in the urban areas. Information from fixed monitoring stations (long-term measurements) are therefore integrated with the ones deriving from mobile station (short-term measurements). Short-term measurements are incomplete and so it is necessary to integrate them with long-term measurements to know the real PM10 level concentrations in the metropolitan area. In this paper we propose a procedure to derive PM10 bi-hour mean levels to complete short time measurements accounting for correlation and time variability between fixed monitoring stations. Referring to bi-hour mean concentration of PM10 in Palermo, Sicily, since 2003 to 2005, we propose the use of a single imputation method (Site-Dependent Effect method, SDEM) developed by the authors in a previous research, in order to impute unobserved values on PM10 bi-hour mean levels in a mobile monitoring station. The goodness of the proposed method is assessed by the coefficient of correlation and an index of agreement. We demonstrate that an adequate use of mobile monitoring stations and a good imputation method allow for a reduction in the number of fixed stations, with a consequent retrenchment of expenses and detection of the point-locations where it is useful to set up new fixed monitoring stations.
Lingua originaleEnglish
Numero di pagine0
Stato di pubblicazionePublished - 2008

Fingerprint Entra nei temi di ricerca di 'Air quality and integration of short-term and long-term pollutant data'. Insieme formano una fingerprint unica.

  • Cita questo