Analysis of spatial covariance structure for environmental data

Formulation and evaluation of environmental policy depend upon a general class of latent variable models known as multivariate receptor models. Estimation of the number of pollution sources, the source composition profiles and the source contributions are the main interests in multivariate receptor modelling. Different approaches have been proposed in the literature when the number of sources is unknown (explorative factorial analysis) and when the number and type of sources are known (regression models). In this paper we propose a flexible approach on the use of multivariate receptor models that takes into account the extra variability due to the spatial dependence shown by the data. The method proposed is applied to Lombardia air pollution data.