Traffic Emission Simulation and Validation with Measured Data in South Korea

Abstract

In major urban areas, air quality is directly related to traffic emissions. Major efforts have been made in recent years to compute emissions from simulated traffic conditions at microscale level in an accurate way. To validate this modelling techniques on-road air pollutant emissions of Nitrogen Oxides (NOx) estimated from mobile monitoring laboratory data are compared to the computed emissions results. The selected location is a traffic hot-spot with more than 4000 vehicles at weekday peak hours. The study area is a 300 m x 300 m domain focused on a signalized intersection of two major roads in Seoul (South Korea). For the traffic simulation, detailed information of the network (geometry, number of lines and traffic light location) is needed along with traffic light cycles. Traffic volumes, routes and fleet composition are estimated from CCTV cameras. Speed-time profiles for peak and off-peak hours on weekdays are simulated with the traffic microsimulation model PTV VISSIM. The predicted profiles are compared to the GPS data measured by the mobile laboratory in real traffic conditions. This comparison is useful to discriminate potential errors from the traffic simulation and the emission computation algorithms. Traffic emissions are computed with VERSIT+micro emission models through the TNO ENVIVER interface on a grid of 5 m x 5 m spatial resolution and temporal resolution of 1 hour along with the emission results of the individual trips. This contribution is useful to validate the emission results computed with the VISSIM-VERSIT+micro/ENVIVER modelling system.