Application of a Multi-Lag Regression Approach to Determine On-Road PM_(10) and PM_(2.5) Emission Rates

摘要

Intensive measurements were performed at roadside locations in Reno, Nevada, and Durham/Research Triangle Park (RTP), North Carolina. Measurements included, but were not limited to, continuous PM concentrations (PM_(2.5), and PM_(10)), wind speed, wind direction, and vehicle number, class, and speed. The results indicated that the observed PM concentration at a given time is a function of PM emissions from vehicles passing the sensors at earlier time intervals. In order to calculate mass emission rates using the continuous data, we applied a multi-lag linear regression approach. The output of the multi-lag regression model, the regression coefficients, along with the vehicle specifications and the wind speed/direction at each site were then used to calculate mass emission factors for passenger cars, light duty trucks, buses, dump trucks, and heavy duty diesel trucks. The chemical mass balance (CMB) was utilized to apportion the mechanisms contributing to the emission factors, including resuspended road dust, tailpipe, and brake wear. This paper reports the results of this study.