Supervision

V It is well known the health effects associated with air pollution. Several studies were published, showing an increasing awareness among the scientific community and policy makers about public health problems due to exposure to air pollution. Policy makers require more detailed air quality information to take measures to reduce the effects on health and to improve the air quality management. This thesis aims to develop procedures using statistical methods to support a broad range of policy decisions concerning this subject. The first aim was the application of statistical methods, such as principal component analysis and cluster analysis, to characterize the air pollution behaviours in an urban area allowing the identification of monitoring sites with redundant air quality measurements. These measurements represent a loss of profitability of the air quality monitoring network resources. Accordingly, the application of these statistical methods let the following policy decisions: (i