For some decades the air quality has received, both locally and globally, more and more attention from the scientific community and local authorities. It is well known that one of the major causes of pollution is vehicle emissions, and that they are closely related with driving style. It has been found that vehicle emissions contain a wide variety of pollutants, principally carbon monoxide and dioxide, oxides of nitrogen, particulate matter and hydrocarbons or volatile organic compounds, which have a major long-term impact on air quality.This paper describes the correlation between the traffic phenomena of road networks and the percentages of the major pollutants made by vehicle emissions. It introduces a new methodological framework application to address the problem of estimating pollutant emissions for large congested urban networks related to real time traffic congestion phenomena. This has been made possible through the studies of impact of signalized intersection, analyzing possible scenarios that lead to decrease of vehicular emissions referring to the French Region of Seine-Saint Denis. Results obtained from a mesoscopic approach offer insights on the use of ITS systems, such as the intelligent traffic lights or Urban Traffic Control (UTC) systems, which are useful to road management policy in terms of pollution decrease. The use of a mathematical model allows the creation of measurement scales of the environmental risk considering signalized and un-signalized intersections. It allows also the comparison of the concentration given by the mathematical modelling fed by real-time data with the European and local Legislation threshold values.