Index of content:
Volume 116, Issue 5, November 2004
- AEROACOUSTICS, ATMOSPHERIC SOUND 
116(2004); http://dx.doi.org/10.1121/1.1804633View Description Hide Description
Noise barriers are commonly used to protect communities from transportation noise. In the present study, three types of barriers, modeled as half planes, have been tested in the laboratory: a conventional rigid barrier with a straight top edge, a straight top edge barrier covered with sound absorbing material, and a rigid barrier with a jagged top edge. Measurements were taken not just behind the barriers, but around them on a plane perpendicular to their top edge. Measured signals were compared against theoretical predictions contributing to further validation of a theoretical model. The sound absorbing material was found to affect the diffracted field more in the front of the barrier than behind it. The diffracted field in front of a jagged edge barrier, similar to the field behind it, was found to depend on the geometry of the edge in the area where the shortest diffraction path intersects the edge profile. Last, the performance of the three barriers was compared with one another in all areas around the barrier. It was found that the jagged edge barrier provides shielding similar to the sound absorbing barrier but at a fraction of the cost.
116(2004); http://dx.doi.org/10.1121/1.1781618View Description Hide Description
In a series of short-range experiments, attempts were made to relate the prevailing meteorological conditions to significant wave shape changes in a 2 ms acoustic pulse propagating over distances from 0.5 to 16 m. Unlike the findings in long-range studies involving time and spatial averages, no meaningful correlations between wind speed and acoustic pulse shape parameters were obtained. Pulse delays could be predicted provided detailed measurements of the wind speed were obtained. Observations showed that two adjacent paths could experience quite different propagation regimes. The geometry and pulse nature of the experiments described here restrict the scale of turbules in any cascade suggesting that a single turbule model may be applicable. Indeed using one or two effective turbules was capable of explaining many of the observations. One model approach used typical parameter values obtained from the literature or from measurements and by locating a single turbule at an appropriate distance from the source receiver axis, correctly generated the observed variety of wave forms. Alternatively, relying on measured data from one distorted pulse and the spatial wind profile, the model was used to determine subsequent pulse distortions. This was successful in predicting behavior in 76% of cases.