Managing data diversity in air quality monitoring and dynamical mapping

In the last decades, air quality monitoring networks have been increasingly installed around the world, with designs which are developed often on a local basis. For example, the European Community gives general rules for the member states which demand local governments to design and manage such local networks. As a result, even if modern instruments are rather precise, the EC monitoring network is very expensive and appears rather etherogeneous from the point of view of spatial representativeness, human risk exposure etc.. Thickening the network at the global scale is an unaffordable task. Satellite measurements are then an interesting data source because of homogeneity over time and space and fixed cost. Along these lines, in this paper, we discuss statistical issues in air quality indexes and spatio-temporal modelling for merging ground level data, computer simulation outputs and satellite data.