Comparison of estimation of emissions based on static and dynamic traffic assignment

For the estimation of air quality and noise nuisance often the output of static traffic assignment models is used. However, static traffic model have several limitations related to the realism of dealing with saturated traffic conditions, which are important for the estimation of emissions. Static traffic assignment models predict congestion on the wrong locations and can not deal with traffic flow phenomena like spillback. As a result also air quality or noise bottlenecks can be predicted on wrong locations, which may lead to erroneous policy decisions concerning new infrastructural measures or implementing mitigating measures on wrong locations. Dynamic traffic assignment models can take these traffic flow phenomena into account and are therefore more suitable to estimate emissions and can in the long term replace the use of static traffic assignment models for larger networks. To provide insights in the differences between estimates based on static and dynamic traffic assignment models in realistic cases, a highway case and urban case are used applying the ARTEMIS model (air emissions) and AR-INTERIM-CM model (noise emissions). The comparison for both cases shows that large differences are found in the estimates for noise (i.e. > 3 dB(A)) and also for the local estimates of air emissions (e.g. up to 45% for NOx emissions). When an infrastructural measure is taken the comparison shows that not only the location and extent of effect, but also the direction of effect can be different (i.e. increase versus decrease and vice versa).