In this paper our objective is to propose a flexible model able to integrate different environmental data subject to heterogeneity. In particular we consider PM10 data coming from monitoring networks for air quality assessment; in this case the heterogeneity can arise because of the different instruments used in the monitoring station and the sampling strategies that change in time and space. In order to obtain comparable data we provide statistical framework able to capture the essential features of the observed spatial-temporal process separating the terms relating to the data heterogeneity. To do this we propose a Geostatistical Dynamical model (GDC) based on the state – space approach introduced by Fassò and Nicolis in [1]; this approach is a geostatistical extension of the DDC model presented in [2]. We assume that the observed data are random fields composed by a linear function of the “true” levels and error components, where the “true” concentrations of PM10 are unobservable processes and represent the state equation of the model. Considering the PM10 data of the Piemonte region, we show some preliminary results.

A statistical approach to heterogeneous monitoring networks

FASSO', Alessandro;NICOLIS, Orietta;CAMELETTI, Michela
2005-01-01

Abstract

In this paper our objective is to propose a flexible model able to integrate different environmental data subject to heterogeneity. In particular we consider PM10 data coming from monitoring networks for air quality assessment; in this case the heterogeneity can arise because of the different instruments used in the monitoring station and the sampling strategies that change in time and space. In order to obtain comparable data we provide statistical framework able to capture the essential features of the observed spatial-temporal process separating the terms relating to the data heterogeneity. To do this we propose a Geostatistical Dynamical model (GDC) based on the state – space approach introduced by Fassò and Nicolis in [1]; this approach is a geostatistical extension of the DDC model presented in [2]. We assume that the observed data are random fields composed by a linear function of the “true” levels and error components, where the “true” concentrations of PM10 are unobservable processes and represent the state equation of the model. Considering the PM10 data of the Piemonte region, we show some preliminary results.
book chapter - capitolo di libro
2005
Fasso', Alessandro; Nicolis, Orietta; Cameletti, Michela
File allegato/i alla scheda:
Non ci sono file allegati a questa scheda.
Pubblicazioni consigliate

Aisberg ©2008 Servizi bibliotecari, Università degli studi di Bergamo | Terms of use/Condizioni di utilizzo

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10446/20594
Citazioni
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact