The aim of this talk is to introduce a method to estimate the production of ozone due to vehicular traffic. The traffic flow is modeled via the second-order Collapsed Generalized Aw-Rascle-Zhang model, choosing a special velocity function and calibrating it on the NGSIM dataset. Then we introduce an emission model to estimate the emission rates of nitrogen oxides (NOx), which are precursors of the ozone in atmosphere. Such a model correlates the emission rate to velocity and acceleration of vehicles. Finally, we consider the photochemical smog cycle, described by three chemical reactions which involve NOx gases and lead to the production of ozone. These reactions are modeled via a system of ODEs, with the source term given by vehicular traffic. We analyze the ozone impact of various traffic scenarios and describe the effect of traffic light timing on such impact.