Estimation of Pollutant Emissions from the Road Traffic at a City Scale, and Its Sensitivity as Regards the Calibration of the Static Traffic Assignment Models

Abstract

In this work, we study how uncertainties on input data of the coupling traffic of the assignment and pollutant emission models affect the resulting traffic and pollutant emissions. We analyse the sensitivity of the approach as regards the scenario definition (O/D matrix and vehicle fleet composition) and as regards the calibration process, which mostly consists in fitting the speed-flow curves. Variations on scenario definition induce changes of about 15% on pollutant emissions and traffic whereas variations on speed-flow curves affect them by about 5%. Complementary analyses show that variation of the travel demand significantly influences traffic, variation of the vehicle fleet composition significantly influences particulate matter (PM), CO2 and NOx emissions, while variation of the shape of speed-flow curve mainly influences the traffic speeds.