Frank A. Boyle

A model for high‐frequency acoustic backscatter from gas bubbles in sandy sediments at shallow grazing angles

A model for acoustic backscatter from trapped bubbles in sandy sediments was developed. The model combines a Biot acoustic penetration model with a resonance scattering mechanism from trapped bubbles. The bubble size distribution is assumed to mirror the size distribution of the fluid pores that exist between sand grains. An estimate of the pore size distribution is constructed from the grain size distribution, based on the known pore structure between dense random packings of hard spheres. The principle of acoustic reciprocity is employed to compute backscattered acoustic pressure in terms of the incident pressure and the scattering cross section of a bubble distribution. The model is applied to data from experiments recently taken at sea. It is concluded that trapped gas is often a likely cause of observed backscatter from sandy sediments. Very small amounts of gas appear to be sufficient to produce significant backscatter.

Nonlinear acoustic scattering from a gassy poroelastic seabed

A model for difference frequency backscatter from trapped bubbles in sandy sediments was developed. A nonlinear volume scattering coefficient was computed via a technique similar to that of Ostrovsky and Sutin [“Nonlinear sound scattering from subsurface bubble layers,” in Natural Physical Sources of Underwater Sound, edited by B. R. Kerman (Kluwer, Dordrecht, 1993), pp. 363–373], which treats the case of bubbles surrounded by water. Biot’s poroelastic theory is incorporated to model the acoustics of the sediment. Biot fast and slow waves are included by modeling the pore fluid as a superposition of two acoustic fluids with effective densities that differ from the pore fluid’s actual density and account for its confinement within sediment pores. The principle of acoustic reciprocity is employed to develop an expression for the backscattering strength. Model behavior is consistent with expectations, based on the known behavior of bubbles in simpler fluid media.

A model for acoustic backscatter from muddy sediments

It is well known that muddy sediments often contain significant amounts of gas of biological origin. Since the scattering cross section of a gas bubble in water is typically 1000 times its geometric cross section, it is reasonable that an acoustic backscatter model intended to work over muddy sediment should contain a bubble resonance scattering component. In this paper a heuristic model is presented, based on scattering from a distribution of suspended bubbles in mud. For acoustic propagation purposes, the mud is treated as a viscous fluid. Since estimates of bubble size distributions in mud are currently unavailable, a bubble distribution similar in shape to that observed in the water column is assumed. The model’s only free parameter is the gas fraction, which can be varied to fit the model to observed data. Small amounts of gas appear to be sufficient to produce observed levels of backscatter. For a homogeneous bubble distribution, the model can be inverted to give an estimate of the gas bubble size d...

Image processing techniques for underwater acoustic image enhancement

Underwater acoustic images are often dominated by noise, which obscures signals of interest. When the spatial character of the noise differs from that of the signal, image processing techniques can be employed to enhance the image. In underwater acoustic time‐bearing displays, the signals of interest form continuous tracks, while noise often takes on the form of isolated speckle. In this case, a speckle suppression technique related to the Hough transform can be used to reduce the noise and enhance the tracks. The presentation will include a discussion of variations of the technique that are under consideration, as well as considerations of optimum values for the input parameters.

Acoustic array data compression via Karhunen‐Loeve transform

Acoustic data compression has been explored in several contexts and several techniques exist. The applicability of each technique depends on the type of data processing that is intended. For example, audio data is often compressed via perceptual coding methods (e.g., mp3) in which quantization noise is distributed according to psychoacoustic principles. The Karhunen‐Loeve (KL) transform presents an opportunity to compress surveillance array hydrophone‐level data while preserving relevant features in beamformed displays. A KL codec was formulated and applied to acoustic test data from a horizontal line array. The results appear promising in that features of interest are preserved with significant data compression. The presentation will include a description of the algorithm as well as examples with actual data. [Work funded by ONR.]

Inferring array geometry from multiple sources of opportunity

A beamformer’s performance is sensitive to the accuracy of the element location estimates. Array element localization is a reasonably simple task when a source is present with a known location. The problem is more difficult with a source whose position is unknown. Techniques commonly require assumptions to be made regarding the source’s range. A method for determining the array’s geometry from multiple sources of opportunity has been formulated. Array geometry is determined from multiple observations of parallax from each source, for a collection of hydrophone pairs in the array. The array’s geometry is built up from the combined orientations of hydrophone pairs. In its current form, the technique requires accurate knowledge of hydrophone pair separation. The presentation will include an application of the method to simulated array data and a discussion of current limitations.

A near field technique for acoustic array element localization

A beamformer’s effectiveness is sensitively dependent on accurate knowledge of the array element positions. Current techniques for using acoustic data to localize array elements are generally quite effective with distant acoustic sources, but are of questionable accuracy when an array must be localized with sources located in the extreme near field. A geometric technique for localizing arrays was developed and appears effective for near field sources. The method was tested with impulsive acoustic data collected from a line array deployed in shallow water, ensonified with light bulb implosions from various positions very close to the array. The presentation will include a discussion of the technique and the experimental results.

Color rendering of range in passive acoustic displays

Techniques for mapping image color to range in passive time‐bearing displays were explored. They included (1) a range‐focusing method wherein images focused at multiple ranges were combined; and (2) a parallax technique wherein images were formed with different subarray apertures. In both cases, a Hough transform based image processing method was used to enhance features and to compensate for degradation associated with combination of multiple images. Colormapping techniques were employed to render source ranges in color. The presentation will include a description of the methods and a discussion of their effectiveness under various conditions. [Work funded under ARL:UT IR&D program.]

An investigation of seabed acoustic penetration and scattering via broadband echo soundings

Recent experiments were conducted in the Mediterranean Sea with the intention of developing a better understanding of the physics of acoustic penetration into and scattering from the seabed. Two shallow water sites were studied, one near Viareggio, Italy and the other off the coast of Sardinia. One site consisted of a fine mixture of silt and sand, with a large amount of trapped gas, while the other included a layer of marine vegetation that produced a bright acoustic return ahead of the bottom echo. Echo soundings were collected with a broad beam impulsive sound source operating over a band between 1 and 15 kHz. The data acquisition system was configured to record the acoustic pulse before and after encountering the seabed, from which a calibrated measure of the seabed’s reflective character was obtained. The measured reflection coefficient was compared with model predictions, based on a poroelastic theory for the seabed. In general the measured reflection coefficients exceeded the predictions significan...

Measuring the locations of penetrating acoustic ray paths into marine sediments

Recent experiments at sea and in the laboratory indicate anomalously high levels of acoustic penetration into sandy marine sediments at subcritical grazing angles, which are not accounted for via current viscoelastic acoustic models for sediments. Current research is directed toward a possible physical explanation. One possibility [Chotiros, J. Acoust. Soc. Am. 97(1), 199–214 (1995)] is that the sediment acts as a Biot poroelastic medium, supporting Biot slow waves with sufficient amplitude to account for the observed penetration. Another [Thorsos et al., SACLANT Conf. Proc. CP‐45, 563–569 (1997)] is that seabed interface roughness scatters a significant amount of sound into the sediment. These two mechanisms differ in their ray geometries for penetrating sound. In order to determine the relative importance of the two mechanisms, a tank experiment was performed to determine acoustic ray paths into a sandy bottom. It involved a narrow beam acoustic source directed at hydrophones in the sediment column. Ini...