version française rss feed
HAL : hal-00273137, version 1

Fiche détaillée  Récupérer au format
Case Studies in Business, Industry and Government Statistics 1, 2 (2007) 119-129
Air pollution sources apportionment in a french urban site
Marie Chavent1, 2, Hervé Guégan3, Vanessa Kuentz1, 2, Brigitte Patouille1, Jérôme Saracco2, 4

The development of air quality control strategies is a wide preoccupation for human health. In order to achieve this purpose, air pollution sources have to be accurately identified and quantified. This case study is part of a scientific project initiated by the French ministry of Ecology and Sustainable Development. Measurements of chemical composition data for particles have been realized on a French urban site. The work presented in this paper splits into two main steps. In the first one, the identification of the sources profiles has been reached thanks to Principal Component Analysis (PCA), followed by a rotation technique. Then, in the second step, a receptor modelling approach (using Positive Matrix Factorization as estimation method) allows to evaluate the apportionment of the sources. The results from these two statistical methods have enabled to characterize and apportion five sources of fine particulate emission.
1 :  IMB - Institut de Mathématiques de Bordeaux
2 :  INRIA Futurs - CQFD
3 :  CENBG - Centre d'Etudes Nucléaires de Bordeaux Gradignan
4 :  GREThA - Groupe de Recherche en Economie Théorique et Appliquée

Pollution data – Principal Component Analysis (PCA) – Positive Matrix Factorization (PMF) – rotation
Liste des fichiers attachés à ce document : 
CSGIGS_chavent_et_al_preprint.pdf(3.8 MB)