Prediction of Pm10 Concentrations at Urban Traffic Intersections Using Semi-Empirical Box Modelling with Instantaneous Velocity and Acceleration
Hong-di He,Wei-Zhen Lu,Yu Xue
DOI: https://doi.org/10.1016/j.atmosenv.2009.09.027
IF: 5
2009-01-01
Atmospheric Environment
Abstract:At urban traffic intersections, vehicles frequently Stop with idling engines during the red-light period and speed up rapidly during the green-light period. The changes of driving patterns (i.e., idle, acceleration, deceleration and cruising patterns) generally produce uncertain emission. Additionally, the movement of pedestrians and the influence of wind further result in the random dispersion of pollutants. It is, therefore, too complex to simulate the effects of such dynamics on the resulting emission using conventional deterministic causal models.For this reason, a modified semi-empirical box model for predicting the PM10 concentrations on roadsides is proposed in this paper. The model constitutes three parts, i.e., traffic, emission and dispersion components. The traffic component is developed using a generalized force traffic model to obtain the instantaneous velocity and acceleration when vehicles move through intersections. Hence the distribution of vehicle emission in street canyon during the green-light period is calculated. Then the dispersion component is investigated using a semi-empirical box model combining average wind speed, box height and background concentrations. With these considerations, the proposed model is applied and evaluated using measured data at a busy traffic intersection in Mong Kok, Hong Kong. In order to test the performance of the model, two situations, i.e., the data sets within a sunny day and between two sunny days, were selected to examine the model performance. The predicted values are generally well coincident with the observed data during different time slots except several values are overestimated or underestimated. Moreover, two types of vehicles, i.e., buses and petrol cars, are separately taken into account in the study. Buses are verified to contribute most to the emission in street canyons, which may be useful in evaluating the impact of vehicle emissions on the ambient air quality when there is a significant change in a specific vehicular population. (C) 2009 Elsevier Ltd. All rights reserved.