Air quality impact assessment of traffic policy in Milan

This paper presents a general spatio-temporal model for assessing the air quality impact of environmental policies which are introduced as abrupt changes. The estimation method is based on the EM algorithm and the model allows to estimate the impact on air quality over a region and the reduction of human exposure following the considered environmental policy. Moreover, impact testing is proposed as a likelihood ratio test and the number of observations after intervention is computed in order to achieve a certain power for a minimal reduction. An extensive case study related to the introduction of the congestion charge in Milan city and the monitoring of particulate matters and total nitrogen oxides motivates the methods introduced and illustrate implementation issues and inferential machinery.