Observation, theory and simulation of anisotropy in oceanic ambient noise fields and its relevance to Acoustic Daylight imaging

Acoustic Daylight is a new technique for creating pictorial images of undersea objects from the acoustic illumination provided by the ambient noise field. As in conventional photography, the directionality of the illumination affects the contrast in the image through the shadows that are cast. Three aspects of the directionality, or anisotropy, of ambient noise in the ocean are discussed in this paper, in the context of Acoustic Daylight imaging: firstly, some observations of the horizontal anisotropy of the ambient noise around Scripps Pier are reported, which indicate that the pier itself is a significant source of noise; secondly, a theoretical model of the acoustic contrast under differing degrees of noise anisotropy is described; and finally, a numerical simulation algorithm that generates Acoustic Daylight images is used to illustrate pictorially the effects of shadowing when the illumination is a representative shallow water noise field. INTRODUCTION For many years, anecdotal evidence has indicated that a submarine can be detected acoustically when it lies above the receiver, because it blocks some of the ambient noise created at the sea surface by breaking waves and related processes. Thus, the submarine casts an acoustic shadow, which in principle is detectable since it contrasts with the background noise field. This phenomenon, as a method of detection, was investigated by Flatté and Munk1 at the suggestion of Dr. Allen Ellinthorpe. At about the same time, Buckingham introduced the idea that, as the ambient noise field is analogous to daylight in the atmosphere, it should be possible to create pictorial images of objects in the ocean through the acoustic "illumination" provided by the noise. If this process could be demonstrated, there would be a substantial impact both on the design and operation of man-made underwater sensing systems, and on our comprehension of the means by which echolocating marine mammals sense their environment. For man to create such an image, a multi-beam acoustic lens is required to focus the noise scattered by the object, along with appropriate signal processing and a visual display monitor to convert the acoustic information into a visual image. The image on the monitor would have much in common with a conventional photograph, although the resolution would be comparatively poor due to the relatively long acoustic wavelengths. The pixels in the image would show a one-to-one correspondence with the beams formed by the acoustic lens; and the angular beam width, which is a measure of the resolution, would be given approximately by the ratio of the wavelength to the aperture of the lens. In the first experiment on Acoustic Daylight imaging, conducted off Scripps Pier in the summer of 1991, a single-beam acoustic lens, consisting of a parabolic reflector with a hydrophone at the focus, was used to identify the presence or absence of rectangular targets mounted on the sea floor2,3. The sole acoustic illumination used in the experiment was ambient noise. Over the frequency band from 5 to 50 kHz, the targets were found to be visible, in that the noise level across the band as perceived at the receiver increased by some 3 dB when the targets were placed in the beam. This result was repeatable over a period of three days, and again three months later, after the experiment had been taken down and reassembled. For various logistical reasons, it was not possible to monitor the directionality of the ambient noise during the first experiment, although it was suspected at the time that the pier is a strong source of noise. Since then, measurements of the anisotropy of the noise field in the horizontal have been conducted, and the original suspicions confirmed: the pier is indeed a strong source of noise, which is both biological (snapping shrimp) and hydrodynamic (wave-breaking around the pilings of the pier) in origin. To investigate the effects of noise anisotropy on Acoustic Daylight images, two different theoretical approaches have been developed. The first is a wave-theoretic model4 in which the target is a pressure-release sphere and the acoustic lens is an endfire line array lying along a radial. An expression is developed in the analysis for the acoustic contrast of the target under differing degrees of noise anisotropy. This expression predicts levels of contrast that are consistent with the original observations from the pilot experiment conducted off Scripps Pier. In the second treatment of the problem, a numerical simulation5, based on Helmholtz-Kirchhoff scattering and a farfield approximation, produces Acoustic Daylight images of objects with smooth curved surfaces. Changes in shadow structure in response to variations in noise anisotropy are qualitatively evident in these simulated images, and again the theoretical results are consistent with the observations from the pilot experiment. In this paper, the three aspects of noise mentioned above are discussed more fully: 1) measurements around Scripps Pier; 2) the wave-theoretic model; and 3) the numerical simulation. The common theme here is the anisotropy in the noise field and the effect that it may have on Acoustic Daylight images. THE NOISE FIELD AROUND SCRIPPS PIER Figure 1 shows an example of the ambient noise time series measured off Scripps Pier in the pilot experiment on Acoustic Daylight. Clearly, the noise is highly non-Gaussian, showing sporadic bursts that are characteristic of biological activity and/or wave action. This is consistent with two facts: the pier and immediate surroundings are known to support populations of snapping shrimp; and the concrete pilings are densely covered with crustacea and molluscs, which give rise to the formation of bubbles after the passage of a surface wave. Visually, this bubble formation process can be correlated with audible airborne sound. To determine the extent to which the pier is a source of noise, underwater acoustic measurements were performed, north and south of the pier, using a directional hydrophone* operating over the frequency band 8 to 80 kHz. The hydrophone was deployed from the stern of an anchored sailing vessel and rotated mechanically in bearing increments of 30 ̊. Figure 2 illustrates the configuration of the experiment as well as the results that were obtained. The polar diagrams, showing the noise intensity as a function of bearing and frequency, are centred on the two stations at which the measurements were made. The noise level shown in each polar plot has been * The hydrophone was provided by courtesy of EDO Corporation, NY. normalized to the beamwidth, to eliminate the effect of the frequency dependence of the latter. Thus, in effect, the noise level in a particular direction shown in Figure 2 is that of the equivalent isotropic noise field as determined by an omni-directional sensor.