Statistical modeling of air pollution

Abstract

The present communication deals with the application of several chemometric methods (principal components analysis, source apportioning on absolute principal components scores, chemical mass balance, self-organizing maps) to various aerosol data collections from different regions in Europe. It is shown that different latent factors explaining over 75 % of the total variance are responsible for the data structure and could be reliable identified and interpreted. Further, the contribution of each identified source to the formation of the particle total mass and chemical compounds total concentration is calculated. Thus, a reliable assessment of the air quality in the respective region is done. Classification by self-organizing maps makes it possible to better understand the role of different discriminating tracers in the air pollution. The use of chemical mass balance approach ensures a sound modeling of the pollution sources. The requirements of the sustainability concept for ecological indicators in this case is easily transformed to a multivariate statistical problem taking into account not separate indicators but the specific multivariate nature of the aerosol pollution.

Keywords:

• Multivariate statistics
• receptor models
• SOM
• aerosol
• air quality.

Acknowledgment

The authors would like to express their sincere gratitude to UNION2 Project (DCVP-02/2-2009) for the financial support.