A Wind Tunnel Study to Examine the Edge Effects of Roadway Barriers

In both field measurements and wind tunnel studies, noise barriers have been demonstrated to generally improve air quality downwind from roadways in most circumstances. However, data are lacking and thus questions remain regarding the specific behavior of the pollutant plume as it disperses around the barrier ends. There is a concern, for instance, that a barrier positioned downwind of a roadway may guide highly polluted plumes from traffic emissions along the barrier leading to heightened concentrations as the plume spills around and downwind of the barrier end. To quantify these “edge effects”, a wind tunnel study was performed to examine concentration patterns downwind of noise barriers with various approach flow wind directions ranging from perpendicular to the roadway to nearly parallel.

The study was conducted in the EPA Fluid Modeling Facility’s meteorological wind tunnel which has a test section measuring 370 cm wide, 210 cm high and 1830 cm long. The study focused on the edge effects of noise barriers, i.e., the manner in which the pollutant emanating from the roadway source is dispersed over and around the end of the barrier. Each roadway segment simulated in the wind tunnel had full-scale equivalent dimensions of 135 m long. A six-meter tall barrier ran the entire 135 m length of the source, located on the downwind side of the source at a distance of 18 m from it (measured perpendicularly from the source). The wind tunnel experiment was performed at a scale of 1:150, making the source and barrier length equal to 90 cm and the barrier height 4 cm. The model roadway consisted of a single line source segment and noise barrier segments arranged in various configurations and at several angles to the wind direction. Pollutant distributions were characterized by measuring a nearly neutrally-buoyant tracer material (ethane) as it dispersed downwind from the line source. Measurements were performed with a bank of six flame ionization detectors digitized at a rate of 20 Hz and averaged over a 120 s period.

The data suggest that while the downwind concentrations of the plume were higher near the ends of the barriers than at the middle of the barrier, those elevated concentrations were similar in magnitude to those measured for the no barrier case, at least at ground level. Above ground level and particularly near the barrier height, the barrier had the effect of mixing the plume vertically which caused higher concentrations than the no-barrier case at all locations along the road segment. The data from this study allow us to quantify these edge effects. In addition to the limited field data available on this issue, this wind tunnel data will be essential for evaluating the ability of noise-barrier algorithms within dispersion models to simulate the overall mitigating effect of barriers.