AERMOD for near-road pollutant dispersion: Evaluation of model performance with different emission source representations and low wind options

- Volume source representation yields better upwind concentrations than area sources.
- All low wind options of volume source representation over-predict upwind concentrations.
- Low wind option 3 is comparatively the best in representing horizontal plume meander.
- Model performance is better near the surface (0.5 m) than that at 3.5 and 9.5 m.

The performance of the regulatory dispersion model AERMOD in simulating vehicle-emitted pollutant concentrations near-roadway using area or volume source representation of emissions and with different low wind options was assessed using the SF6 tracer data from the General Motors (GM) Sulfur Dispersion Experiment. At downwind receptor locations, AERMOD, using either area or volume source emissions, can reasonably predict the tracer concentrations near the surface (0.5 m) but the model performance decreases at higher elevations (3.5m and 9.5m above the surface). For upwind receptors, using an area source representation leads to significant under-predictions due to AERMOD's lack of treatment of lateral plume meander, but using volume source representation leads to over-predictions of upwind concentrations regardless of the low wind options for plume meander. Among the three low wind options currently available in AERMOD, best model performance is obtained with low wind option 3, which treats plume meander with a higher minimal standard deviation of the horizontal crosswind component (σv,min = 0.3 m s−1), eliminates upwind component of dispersion and uses an effective lateral dispersion parameter (σy) to replicate centerline concentration. The optional adjustment of the surface friction velocity in the meteorological preprocessor AERMET does not lead to obvious improvements in predicted near-road concentrations for this application.