On-road measurements and modelling of vehicular emissions during traffic interruption and congestion events in an urban traffic corridor

Traffic events such as interruptions and congestions occur on urban roads causing higher vehicular emissions resulting in air-pollution hotspots. This study investigates the performances of two emission models for estimating emissions from passenger cars and auto-rickshaws of different mileages, moving with a traffic fleet during these events. The instantaneous measurements of emissions and speeds were carried out on a test route of 3.8 km by integrating the auto-gas analyser and V-Box for estimating the on-road emission factors (EFs). The measured driving profiles and acceleration/decelerations were used in the IVE emission model and the speed was averaged in the COPERT-IV emission model into 0-10 km/h; 10-25 km/h; 25-35 km/h and > 35 km/h. The on-road EFs of CO, HC, CO2 and NOx at peak hours (PHs) and off-peak hours (OPHs) were compared with the modelled EFs. The estimates of COPERT-IV were about 25% less for CO, HC and CO2 and 26-39% more for NOx than IVE during PHs. As compared to the on-road EF, the IVE estimates for CO, HC, CO2 was 5-50% less and 12-50% higher for NOx for auto-rickshaw and passenger car; and the COPERT-IV estimates for CO, HC, CO2 were 30-74% less and 6-86% higher for NOx, showing that instantaneous driving profiles in emission modelling are important. The interruptions during PHs were 80% of the time, a cause of deviations in the models. Thus, the real-world emissions vary with the levels of interruptions and congestions on the roads, and hence emission models must account for these events.