Modeling the short-term effect of traffic on air pollution in Torino with generalized additive models

Vehicular traffic typically plays an important role in atmospheric pollution. This is especially true in urban areas, where high pollutant concentrations are often observed. In this paper, we consider hourly measures of concentrations of nitrogen oxides (NO, NO2 and NOx), carbon oxide (CO) and particulate matter (PM), collected at the stations distributed throughout the city of Turin. To help explain the short-term behavior of the concentrations of these pollutants, we propose using generalized additive models (GAM), focusing in particular on traffic along with the meteorological predictors. All the data are collected during the period from December 2003 to April 2005.