Bayesian spatio–temporal modeling of urban air pollution dynamics

This work deals with the spatio-temporal analysis of urban air pollution dynamics from the town of Perugia, Italy, using high-frequency and size resolved data on particular matter. Hierarchical Bayesian models are used that allow for an autoregressive term in time. Some preliminary results show that there is a significant spatio–temporal structure with a large first–order temporal correlation coefficient. Future analysis will concern the use of higher–order temporal auto–correlation structures and the introduction of the effect of some covariates.