Evaluating near-roadway concentrations of diesel-related air pollution using RLINE

The near-roadway pollutant dispersion model RLINE was evaluated for prediction of nitrogen oxides (NOx) and black carbon (BC) concentrations. Model predictions were compared with continuous, yearlong measurements from two near-roadway sites in the San Francisco Bay Area. Heavy-duty diesel trucks were a significant source of NOx and BC at both sites. Characterization of temporal variations in heavy-duty truck activity on diurnal, weekly, and seasonal scales were included in this study; truck traffic and emissions are not well-correlated with passenger vehicle or total traffic volumes. For both pollutants, more than 90% of predicted 24-h average concentrations were within a factor of two of observations at both near-roadway monitoring sites. The model responds appropriately to seasonal variations in meteorology and day-of-week variations in emissions. RLINE model performance for NOx was better overall than for BC. Reducing uncertainties in emission factors would help to improve model performance for BC.