Volume 109, Issue 1, January 2001
Index of content:
- AEROACOUSTICS, ATMOSPHERIC SOUND 
109(2001); http://dx.doi.org/10.1121/1.1328793View Description Hide Description
The amplitude and waveform shape of atmospheric acoustic pulses propagating horizontally over a seasonal snow cover are profoundly changed by the air forced into the snow pores as the pulses move over the surface. This interaction greatly reduces the pulse amplitude and elongates the waveform compared to propagation above other ground surfaces. To investigate variations in snow-cover effects, acoustic pulses were recorded while propagating horizontally over 11 different naturally occurring snow covers during two winters. Two inversion procedures were developed to automatically match the observed waveforms by varying the snow-cover parameters in theoretical calculations. A simple frequency-domain technique to match the dominant frequency of the measured waveform suffered from multiple solutions and poor waveform matching, while a time-domain minimization method gave unique solutions and excellent waveform agreement. Results show that the effective flow resistivity and depth of the snow are the parameters controlling waveform shape, with the pore shape factor ratio of secondary importance. Inversion estimates gave flow resistivities ranging from 11 to 29 kN s m−4, except for two late-season cases where values of 60 and 140 were determined (compared to 345 for the vegetation-covered site in the summer). Acoustically determined snow depths agreed with the measured values in all but one case, when the depth to a snow layer interface instead of the total snow depth was determined. Except for newly fallen snow, the pore shape factor ratio values clustered near two values that appear to correspond to wet (1.0) or dry (0.8) snow.
Perturbation theory applied to sound propagation in flowing media confined by a cylindrical waveguide109(2001); http://dx.doi.org/10.1121/1.1331676View Description Hide Description
First-order perturbation theory is employed to examine sound propagation in flowing media confined by a cylindrical waveguide. The use of perturbation theory allows examination of mode phase-speed changes due to any radially dependent flow as long as the flow magnitude is sufficiently small. The condition to be fulfilled is satisfied in the flow range: m/s for the specific values of cylinder radius, ultrasound frequency, and sound speedanalyzed in the present work [in the general case, however, the condition in Eq. (1) of the present work must be fulfilled]. This freedom of choice, i.e., the possibility to handle any radial flow profile, is used to analyze two flow profile cases: (1) where is a linear combination of a laminar flow profile and a flat profile corresponding to turbulent flow, and (2) where is a linear combination of a laminar flow profile and a more realistic logarithmic-dependent turbulent flow profile. In both cases, it is shown that large errors may result in ultrasoundflow measurements if several modes are excited by the transmitting transducer, and that a logarithmic flow profile in the turbulent regime leads to somewhat larger measurement errors at high flow values as compared to assuming a simple flat profile in the turbulent regime.
Aeroacoustics of diffusers: An experimental study of typical industrial diffusers at Reynolds numbers of109(2001); http://dx.doi.org/10.1121/1.1329618View Description Hide Description
Diffusers as used in gas transport systems have an optimal pressure recovery but are unstable due to marginal flow separation. Coupling of diffuser flow oscillation with acoustic modes in a pipe has been demonstrated in a recent work by Kwong and Dowling [J. Fluids Eng. 116, 842 (1994)] to drive flow unsteadiness. Considered here in addition to the diffuser at a pipe termination is the aeroacoustic response of a diffuser in a long pipe. In both cases reflection coefficientmeasurements show that at moderate and low amplitudes of the acoustical particle velocity compared to the main flowvelocity, diffusers are aeroacoustic sources similar to the whistler nozzle and the horn. This confirms the observations of Kwong and Dowling. At higher acoustical velocity amplitudes diffusers become strong absorbers, which can be explained in terms of a quasistationary flowmodel. Finally, an indication is provided for possible remedial measures when a stable flow is needed.
Numerical modeling of the spectral broadening of sodar echoes by winds perpendicular to the axis of a finite beamwidth antenna109(2001); http://dx.doi.org/10.1121/1.1331677View Description Hide Description
A simple model for determining the amount of spectral broadening in acoustic sounderechoes caused by winds traveling perpendicular to the antenna beam is presented. Key features of the model are that the wind profile is described by an analytic function and the antenna radiation pattern is included in the calculations. The model was restricted to the situation where the antenna was aligned in the vertical direction and the winds were horizontal. The effects of refraction by temperature and wind velocity gradients were neglected. The radiation pattern used in the model was the or function. The wind profiles which were used were a constant windspeed between the antenna and the scattering height, a profile where the windspeed increases linearly with height, and a logarithmic profile. Here, the amount of spectral broadening or spectral width of the echo is calculated by treating the amplitude spectrum of the echo as a probability distribution and determining the standard deviation of this distribution. It was found that the relationship between the standard deviation of the echo, σ, and windspeed, u, was closely described by a linear relationship, with the value of B ranging from 0.05 to 0.08 for the situations modeled here. Observational data are presented. The data were obtained with a sounder whose radiation pattern was approximated with the sinc function. The observed relationship between σ, and u, was essentially linear, agreeing favorably with the model predictions; however, the value of the proportionality constant B was 0.04. A possible explanation for this is that the radiation pattern for the antenna may have been underestimated since the effects of the sound absorbing shielding surrounding the antenna are unknown.