Normal incidence reflection loss from a sandy sediment

Abstract

Executive Summary:The detection of mines buried in the seabed is of continuing concern. The optimization of sonars for such purposes depends critically on the understanding of the interactions of acoustics with the seabed. A well advertized limitation to search geometries is the existence of a critical angle for the incidence of an acoustic wave on the seabed. Beyond this angle most of the acoustic energy is reflected back into the water. In practice, this limitation is ill-defined because of small scale sediment variability and water-sediment interface roughness. There have been a number of experimental and theoretical attempts to further understand the nature of the critical angle limitation. There are models of varying complexity, which may be used to describe the acoustic properties of the sediment. This report provides, through the analysis of a simple normal incidence experiment, evidence to support the appropriateness of one model over another. Abstract:Acoustic reflection loss at normal incidence of a sandy sediment, in the Biodola Gulf on the north side of the island of Elba, Italy, was measured in the band 8 -17 kHz, using a self-calibrating method. The water depth was approximately 11 m the sand pure with a mean grain diameter of 0.2 mm. The measured reflection loss 11 dB, ±2 dB is consistent with measurements in the published literature. The computed reflection loss for an interface between water and a uniform visco-elastic media with the same properties was 8 dB, ±ldB. The theoretical and experimental values do not significantly overlap, which leads to the conclusion that the viscoelastic model is inappropriate. The Biot model is suggested as a better alternative but more work is needed to ascertain the appropriate parameter values.

Experimental Study of Sound Waves in Sandy Sediment

“…A Z E of zero is used in the body of this report. Pressure coefficients computed using Johnson and Plona's method [20] are calculated below and compared to pressure coefficients from the lumped parameter method, (2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19), are calculated and compared to modes shapes from the lumped parameter method, eqs. and .…”

“…The fast and slow effective density numbers in eq. (4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14) are at 250 kHz. As can been seen, the density of the fast wave is 89% + 3% of the total density.…”

Normal incidence measurement in a subaqueous sand dune field in the South China Sea

“…The EDF model is a fluid approximation of the Biot model that shows high accuracy in estimating geoacoustic parameters of sediment and has been extensively investigated (Holland and Brunson, 1988;Chotiros et al, 2002). The inversion technique used in this paper estimates the properties of surficial sediment from normal incidence reflection coefficients, and the procedure closely follows that of Schock (2004) and Chiu et al (2014).…”

“…According to Chotiros et al (2002), seafloor roughness and bottom curvature affect the normal incidence reflection measurement, and these effects can be reduced by averaging the reflected values over several pings. However, for significant bedform curvature, within one acoustic footprint, the normal incidence signals could be scattered or focused, which results in a reduction or an enhancement, respectively, of reflection strength and produces errors in reflection coefficient estimates.…”

Mean grain size mapping with single-beam echo sounders

“…3 This frequency dependence becomes manifest if the product of wave number and transition layer thickness is of order unity or higher. 6 Ainslie 7 summarizes measurements and shows that a more precise condition for the onset of frequency-dependent effects is for the product of acoustic frequency and layer thickness to exceed about 200 m / s. For example, Lyons and Orsi 4 suggest that frequency-dependent measurements 8,9 of normal-incidence reflection loss from a sand sediment may be caused by a strong density gradient at the water-sediment interface. Knowledge of the thickness of the transition layer is required to account for the impedance gradient in modeling the reflection coefficient.…”