Search

Acoustic reflection coefficients are reported for water-saturated granular media at frequencies from 1.2 to 2.0MHz using a narrow-beam broadband transducer in a monostatic geometry at near-normal incidence. Natural sand and glass beads with median grain diameters ranging from 0.22 to 0.40mm were used. For each granular medium, bed elevation and root-mean-square roughness were measured using side-on photographs of the sediment-water interface. The probability density distributions of the bed elevations are Gaussian. The roughness parameter is close to 1, indicating that the reflected pressure field is mainly due to coherent scattering. The probability distribution of the observed reflection coefficients is nearly Gaussian, consistent with the predictions from a coherent single-scattering model. The horizontal decorrelation length of the observed reflection coefficients is˜4 mm, with no consistent dependence on either frequency or grain size, and approximately equal to 20% of the transducer diameter. This behaviour, which is reproduced by the single-scattering model, is due to speckle. The size/frequency-dependence of the reflection coefficients are well described by Eckart’s [(1953). J. Acoust. Soc. Am. 25(3), 566–570] prediction for a rough surface with Gaussian-distributed surface elevations. Comparisons are made to previously reported reflection coefficient measurements. [ABSTRACT FROM AUTHOR]

Measurements of the phase velocity of compressional sound waves in water-saturated granular materials are reported for the 1.0–2.0 MHz frequency range. The sound speed estimates are based on travel times through granular layer thicknesses ranging from 8 to 17 mm. Three types of granular media were used: 336 μm median diameter glass beads and two natural sands with median diameters of 219 and 406 μm. These grain sizes and frequency range correspond to 0.5 < k a < 1.2 , where k is the wavenumber and a the grain radius. To remove trapped air, the samples were boiled under pressure before transfer to the measurement tank. The results are compared to previously reported experimental results and to the Schwartz and Plona [J. Appl. Phys. 55(11), 3971–3977 (1984)] multiple scattering prediction, confirming negative dispersion for ka > 0.5. Scaling the data by a factor depending on porosity and grain density reduces the spread among the available phase speed estimates by nearly a factor of 2, from 12.5% to 6.9%. [ABSTRACT FROM AUTHOR]

Attenuation and group speed measurements are reported for water-saturated granular materials (natural sand and glass beads) at frequencies of 1.0 to 1.8 MHz. Median grain diameters were 219 to 497 μm, corresponding to k d ≳ 1 , i.e., the scattering regime. The measurements were made for different thicknesses of sediment resting on a reflective surface using a monostatic geometry. The attenuation estimates compare well with previously reported experimental results and to the predictions of multiple scattering theory, confirming in particular the tendency toward f4 dependence for k d ≳ 1. Group speed estimates exhibit the negative dispersion predicted by theory and are comparable in magnitude to previously reported measurements made using transmission geometries. It is found that the available data exhibit a O(10)% spread among the sound speed measurements at a given kd value, and that this spread is reduced to 2.2% when the data are scaled by a factor dependent on porosity and grain density, and that essentially all of the reduction can be attributed to differences in porosity. [ABSTRACT FROM AUTHOR]

The short-pulse acoustic backscatter amplitude calibration technique of Dragonette, Numrich, and Frank [(1981), J. Acoust. Soc. Am. 69, 1186–1189] is evaluated for MHz frequency systems, as are used to measure environmental suspended sediment concentration. The technique involves measuring backscatter from a short transmit pulse reflected by a solid spherical target, and treating the first received reflection as equivalent to that from a perfectly rigid sphere. In this implementation, the target sphere is 16mm diameter tungsten carbide. The target is mounted on a stiff support rod rather than being suspended on a filament, enabling precise positioning of the sphere relative to the transducer. That the support rod does not significantly influence the calibration is demonstrated by reconstructing the backscatter time series, including the sequence of pulses from surface (Rayleigh) waves, using elastic scattering theory. Calibration results are verified in a separate experiment in which the backscatter from dilute aqueous suspensions of 400–600 lm diameter polystyrene beads in water is measured within the 1–2MHz frequency band. After correcting for the additional attenuation within the scatterers, the resulting values of the backscatter form factor are within 20% of the theoretical predictions. [ABSTRACT FROM AUTHOR]

A methodology is developed and tested for simultaneously extracting time-resolved one-dimensional profiles of the mass-concentration and velocity of two different particle types in a mixed suspension, using a multi-frequency pulse-to-pulse coherent Doppler instrument. The technique involves inversion of a model for frequency-dependent acoustic backscatter amplitude and phase. Results are presented from a laboratory settling column experiment, measuring a mixture of polystyrene beads (slowly-settling, strongly-scattering) and glass beads (quickly-settling, weakly-scattering) in a vertical pipe section. [ABSTRACT FROM AUTHOR]