Volume 88, Issue S1, November 1990
Index of content:
- PROGRAM OF THE 120TH MEETING OF THE ACOUSTICAL SOCIETY OF AMERICA
- Session 1OC: Acoustical Oceanography: Active Bubbles at the Ocean Surface
- Invited Papers
88(1990); http://dx.doi.org/10.1121/1.2028896View Description Hide Description
It is now quite clear that vast number of microbubbles are produced by breaking waves and precipitation, and that they start their lives as shock‐excited, damped, nonlinear, pulsating sound sources. These screaming infant microbubbles are active for only milliseconds and are, therefore, very close to the oceansurface while they radiate. In the presence of a rough oceansurface, their radiation patterns may appear to come from tilted dipoles or even monopoles, depending on the ocean wave spectrum, the frequency of the sound radiation, and the listening angle. In the absence of rain, it is those infant bubbles created by breaking waves that are responsible for the Knudsen sea noise spectrum from 500 Hz to 20 kHz or beyond. Those active bubbles have been used to determine the oceansurface bubble density, as a function of bubble radius, caused by intermittent spilling breakers during low sea states. Another major component of underwater sound, the infant bubbles produced by rainfall, show promise as keys to the remote determination of the rainfall spectrum (number of drops as a function of the drop diameter) at sea. On the other hand, the passive adult microbubbles, whose highly variable numbers must also depend on the dissolved organic and detrital character of the local ocean, are excellent tracers of near‐surface ocean processes. Some recent laboratory and ocean experiments that lead to those understandings will be reviewed. [Research supported by ONR.]
88(1990); http://dx.doi.org/10.1121/1.2028897View Description Hide Description
The rate of damping of bubble oscillations provides us with a clue as to their precise origin and mode of generation. Data on rates of damping of acoustical pulses from various sources will be examined, particularly those from low sea states. At a given frequency, the observed rates vary by a factor of order 10; the minimum damping rate corresponds closely to thermal damping. Theoretical reasons for the higher damping rates will be discussed, including the transfer of some energy to shape oscillations.
88(1990); http://dx.doi.org/10.1121/1.2028945View Description Hide Description
The observation that oceanic ambient noise is wind dependent over a broad frequency range [G. Wenz, J. Acoust. Soc. Am. 34, 1936 (1962)] suggests a major role for gas bubbles as sources of this sound.Bubblenoise generation mechanisms in a variety of modalities and under both passive and active surveillance are examined. The preliminary studies have examined topics such as collective oscillations of bubbleclouds,bubble entrainment by capillary waves, and precipitationnoise. A brief overview of this research is presented and the most recent results are described in some detail. [Work supported by ONR and ONT.]
88(1990); http://dx.doi.org/10.1121/1.2028946View Description Hide Description
The physics of sound generation from light and heavy rain will be discussed in the context of the measurement of rainfall rate by monitoring underwater ambient sound levels. The physics of sound generation by small drops/light rain at normal and oblique incident will be reviewed. These results will be contrasted with preliminary field measurements of the sound produced by heavy rain during the Acoustic Rain Experiment at the Ocean Test Platform in the Gulf of Mexico. The underwater soundspectragenerated by heavy rain will be explained using new laboratory measurements of the soundgenerated by individual large drops showing acoustic energy from initial impacts (at 10–15 kHz) and from entrained bubbles at lower frequencies.
- Contributed Paper
88(1990); http://dx.doi.org/10.1121/1.2028947View Description Hide Description
Drop diameters from 2.7 to 4.5 mm are common in heavy rainfall. By using a hydrophone with a flat response up to 300 kHz and A/D conversion up to 1 MHz the components of the sound radiation for these large terminal‐velocity drops in the time and frequency domains can be identified. At these high sampling rates it is possible to see notches on the lagging edge of the impulse trace. These notches at 100 kHz and higher frequencies are probably caused by internal drop reflections. Lower frequency oscillations with a range of frequencies less than 10 kHz occur after a time lag of 30 ms or more. Both the frequency of oscillation and the time lag appear to be the functions of drop size. Although the amplitudes of these components are comparable for individual drops the energy radiated by the low‐frequency oscillation is greater due to its longer duration. These early results suggest that it may be possible to ascertain the drop spectrum (number of drops as a function of diameter) in rainfall from the spectrum of the underwater sound of the rain. [Research supported by ONR.]
- Invited Papers
88(1990); http://dx.doi.org/10.1121/1.2028948View Description Hide Description
The soundgenerated by breaking surface waves can be used to track their location, lifetime, and motion across the oceansurface. A six‐hydrophone array of span 8.5 m and 5‐kHz bandwidth has been developed. It forms part of a freely drifting and self‐contained instrument that also includes various active acoustical systems for studying bubbleclouds; it was first used in the Surface Wave Processes Project in February/March 1989. Preliminary results will be presented, focusing on the pattern and behavior of the breaking in different ambient wave environments.
88(1990); http://dx.doi.org/10.1121/1.2028989View Description Hide Description
An elegant series of experiments was conducted recently by Farmer and Vagle [J. Acoust. Soc. Am. 86, 1897–1908 (1989)], in which the sound from individual wave‐breaking events was recorded by a hydrophone located a few meters beneath the oceansurface. Between 1 and 20 kHz, the observed spectra from La Perouse Bank contain relatively sharp, isolated peaks (hot spots) at intervals of approximately 3 kHz, whereas the data set from the FASINEX experiment shows energy spread over broad spectral bands (supermodes), perhaps 3 kHz wide, and separated by fairly well‐defined energy gaps (i.e., spectral minima). Both types of spectrum are a manifestation of waveguide propagation in the upward refracting, ocean‐surface bubble layer. A new, modal theory of sound in an upward refracting duct shows that, in general, the mechanism responsible for the observed hot spots and supermodes is intermode interference. The qualitative differences between the two data sets arise because the number of modes supported by the profile at La Perouse is significantly less than the number in FASINEX. In both cases, the detailed features displayed by the theoretical and experimental spectra show an almost one‐to‐one correspondence. [Research supported by ONR.]
- Session 1ID: Tutorial on Active Control of Sound and Vibration
88(1990); http://dx.doi.org/10.1121/1.2028990View Description Hide Description
Due to recent advances in DSP technology, active control now shows much promise for reducing sound and vibration in a number of difficult applications. In this tutorial the concepts behind the active control approach will be introduced with some historical perspective. The basics of feedforward control will then be reviewed and related to the more traditional state feedback methods. Application of the technique to fundamental problems in the control of sound and vibration will be given with some consideration of the design and choice of the appropriate control transducers and cost functions. Useful analytical methods and mechanisms of control will be considered. Recent successful applications of the technique to realistic systems will be reviewed and future promising research directions discussed.
- Session 2AB: Animal Bioacoustics: Animal Bioacoustics I
- Contributed Papers
88(1990); http://dx.doi.org/10.1121/1.2029029View Description Hide Description
A sperm whale neonate, Physeter spp., was stranded near Sabine Pass, Texas in September of 1989. The Marine Mammal Stranding Network of Texas A&M University removed the animal to Sea‐Arama Marineworld, Galveston, Texas where it was kept in a large tank for 8 days in an attempt at rehabilitation. During this period, the auditory brainstem response (ABR) was recorded from suction cup sensors on the surface of the head in response to pulses presented (20 and 40 pps) underwater near the right external auditory meatus and lower jaw. These are the first ABR records from a great whale and the first recordings from a neonate of any cetacean species. The ABR waves of this Physeter, in response to pulses ranging in peak frequency from 2.5 to 60 kHz, were similar in appearance to those seen in other mammals, and very similar to those previously observed in other odontocetes [S. H. Ridgway et al., Proc. Natl. Acad. Sci. USA 78, 1943–1947 (1981)]. The ABRs of highest amplitude were those to frequencies of 5, 10, and 20 kHz. Responses to 60 kHz were much weaker but definitely present. Latencies were prolonged compared with ABRs of other juvenile and adult odontocetes.
88(1990); http://dx.doi.org/10.1121/1.2029030View Description Hide Description
The site and physiologic mechanism(s) responsible for the generation of odontocete sonar signals have eluded investigators for decades. Examination of postmortem odontocete heads with medical imaging devices (x‐ray CT and MRI) across diverse taxa has identified a complex of structures that may function as the essential components of a biosonarsignal generator. Each monkey lips/dorsal bursae complex (MLDB) is associated with one (right and left) air passage. Each complex is composed of at least two fatty bursae that are embedded in the monkey lips along the airway, a cartilagenous “stiffening” rod, and a stout ligament. The size of the fatty bursae varies with the species but tends to suggest a relationship with wavelengths for peak frequencies of published signals. In addition, the location of these bursae makes them excellent candidates for sonar signal transducers. The details of this complex and its morphologic surroundings will be shown in examples from six extant odontocete families. Preliminary results from an endoscopic investigation of an echolocating dolphin support the hypothesis that this anatomic complex is the site of sonar signal generation.
88(1990); http://dx.doi.org/10.1121/1.2029031View Description Hide Description
It has been established that some dolphins possess well‐developed acoustic orientation (echolocation) and information gathering abilities, though substantially less is known about the system of sound generation and beam formation. Dolphins use a narrowly focused sound beam that emanates from the forehead and rostrum during echolocation. The primary objectives of this study were to simulate the effects of modeled tissues on beam formation, and to test the viability of various hypothetical sound source locations. 2‐D simulations of sound propagation using parasagittal outlines from reconstructions of CT scans through the forehead tissues of two delphinids were conducted. Finite difference wave propagation programs were run on a Cray supercomputer. Preliminary simulations suggest that beam formation results primarily from reflection off of the skull and air sacs. These results do not depend strongly upon the precise values of velocity and density assumed for the bone. Beam angles closely approximate those measured by experimental methods for a source located in a region of the model corresponding to the monkey lip/dorsal bursae complex (MLDB), and not elsewhere. These results suggest that: (1) the skull and air sacs play the central role in beam formation; (2) the geometry of reflective tissue is more important than the exact acoustical properties assumed; and (3) experimentally observed beam patterns are best reproduced in the simulations when the sound source is placed in the region of the dolphin's head known as the MLDB (Cranford, 1988).
88(1990); http://dx.doi.org/10.1121/1.2029032View Description Hide Description
The capability of an Atlantic bottlenose dolphin to discriminate wall thickness differences of hollow cylinders by echolocation was studied. A standard cylinder of 0.635‐cm wall thickness was compared with cylinders having wall thickness that differed from the standard by ±0.2, ±0.3, ±0.4, and ±0.8 mm. All cylinders had an o.d. of 3.81 cm and a length of 12.7 cm. The dolphin was required to station in a hoop while the standard and comparison targets, separated by an angle of ±11° from a center line, were simultaneously presented at a range of 8 m. The dolphin was required to echolocate and indicate on which side of the center line the standard target was located. Target location on each trial was randomized. The dolphin could discriminate wall thickness differences of −0.23 and +0.27 mm at the 75% correct response threshold. Results of backscatter measurements suggest that if the dolphin used time domain echo cues, it may be able to detect difference in the time between two echo highlights to within approximately ±500 ns. If the dolphin used frequency domain cues, it may be able to detect frequency shifts as small as 3 kHz in broadband echoes having a center frequency of approximately 110 kHz.
88(1990); http://dx.doi.org/10.1121/1.2029080View Description Hide Description
An active wide‐beam sonar system for obstacle localization in 2 D is described. Efficient localization is achieved by mimicking the multiple sensor configuration of bats and employing an improved method for time‐of‐flight (TOF) estimation. Three transducers are employed, with a transmitter in the middle (the mouth) flanked by two receivers (the ears). After the transmitted pulse encounters an obstacle, the TOF information is extracted from the noisy echoes detected by each receiver, to estimate the range r and azimuth θ of an obstacle located within the active region of the sonar system. Simple thresholding does not produce accurate results due to the inevitable bias error in the TOF measurement. An unbiased estimate of TOF is developed by modeling the leading edge of the echo waveform by a parabola, whose vertex corresponds to the actual TOF. Based on this TOF information, unbiased range and azimuth estimates are derived for an isolated obstacle. Experimental results show that the estimates are most reliable if the obstacle is located along the system line‐of‐sight and that they improve with decreasing range, providing a novel interpretation for bat foraging. Experimental results indicate only a slight degradation compared to the Cramér‐Rao lower bounds for estimator variances. [Work supported by NSF:ECS‐8802627.]
A method for ranking the acoustic disturbance potential of noise sources in the environment of marine mammals88(1990); http://dx.doi.org/10.1121/1.2029081View Description Hide Description
A standardized noise contribution model (SNC) has been developed to provide a means of comparing the contributions from diverse noise sources to facilitate assessment of their potential environmental impact. The output of the SNC model is a logarithmically scaled number proportional to the acoustic energy density produced by a specific type of source operating in a defined reference area. An associated standardized exposure rating model (SER) has been developed to rate the potential response of marine mammals to noise exposure. The SER model is designed to evaluate the degree of potential impact of a specific source on a specific species by producing a log‐scaled number proportional to the degree of matching between a noise source output bandwidth and a species hearing sensitivity characteristic. The SNC value for a given source type is used together with information on population density to weight the SER values for specific source and species encounter probabilities. [Work supported by the Minerals Management Service.]
88(1990); http://dx.doi.org/10.1121/1.2029082View Description Hide Description
The temporal and spatial patterns in the distribution of animals are poorly understood primarily because of inadequate observation techniques. A new passive localization technique (acoustic tomography) is demonstrated that automatically localizes the positions of calling animals with rms errors of a few centimeters from measurements of the call's travel‐time difference at several pairs of microphones [J. L. Spiesberger and K. M. Fristrup, Am. Naturalist 135, 107'153 (1990)]. Tomographic localization is significantly more precise (by factors from about 2 to 100) than passive acoustic techniques used previously because tomography accounts for and estimates the sound speed and wind fields as well as the positions of the microphones all of which significantly modify the travel‐time differences. Computers are now fast and inexpensive enough for automatic localization of calling animals in large regions of forests and other terrestrial environments.
20‐Hz pulses and other vocalizations of fin whales, Balaenoptera physalus, in the Gulf of California, Mexico88(1990); http://dx.doi.org/10.1121/1.2029083View Description Hide Description
Low‐frequency vocalizations were recorded from fin whales, Balaenoptera physalus, in the Gulf of California, Mexico (mainly the Sonoran coast) during three cruises in March 1985, and February and August 1987. The predominant sounds were patterned 20‐Hz pulses. In March 1985, they were in sequences characterized by regular interpulse intervals of 9 s, whereas in August 1987, they were mainly in sequences of doublets in alternating 5‐ and 17‐s interpulse intervals. In February 1987, no patterned pulse sequences were detected. The 20‐Hz pulses were about 0.5‐s duration, typically sweeping from 42 to 20 Hz. Other fin whale vocalizations were of similar duration and typically modulated downward in frequency, averaging 82, 56, and 68 Hz, respectively, for the three cruises. Gulf pulses differed from those elsewhere in terms of frequency sweep, sound level, and time interval patterns. This uniqueness points out the possibility that fin whales in the Gulf of California may represent a regional stock based on their soundcharacteristics.
88(1990); http://dx.doi.org/10.1121/1.2029119View Description Hide Description
A sperm whale neonate, Physeter spp., was stranded near Sabine Pass, Texas in September 1989. The Stranding Network of Texas A&M University removed the animal to a large tank 6×12.2 m in Galveston, where it was nursed for 8 days. During this period, sounds were recorded with a broadband Racal recorder and two matched broadband B&K hydrophones placed 1 m ahead of the blowhole and 1 m lateral to the fight eye. Sounds could be divided into two classes. The click class consisted of clicks of two different types: (a) high‐frequency, low‐amplitude clicks with peak frequencies of 5 to 12 kHz and 1‐ to 2‐ms duration; (b) low‐frequency, high‐amplitude clicks with peak frequencies of 400 to 900 Hz and durations of 7 to 20 ms. The second or grunt class consisted of different sounds, mostly below 700 Hz, described by listeners as “croaks,” “growls,” “grunts,” and “‘lion‐purr growls.” Palpation of the head during phonation, listening with a stethoscope and in air at mid‐head, and arrival times at the two hydrophones, suggested that click‐class sounds were produced near the blowhole (distal sac and museau) while the grunt class originated 60 to 70 cm posterior to the blowhole (frontal sac).
- Session 2EA: Engineering Acoustics: Flow Noise and Hydrophone Arrays
88(1990); http://dx.doi.org/10.1121/1.2029120View Description Hide Description
The noise induced by the turbulent boundary layerpressurefluctuations on hydrophones embedded in an elastomer layer is usually predicted by assuming that the layer is unbounded in the lateral direction and the hydrophone does not influence the stress field [S. H. Ko and H. Schloemer, J. Acoust. Soc. Am. 85, 1469–1477 (1989)]. These predictions neglect the flow‐induced noise associated with the edges of the layer as well as the diffraction, reflection, or shadowing of the stress field by the hydrophones. To avoid these limitations, the finite element code ATILA [B. Hamonic et al., J. Acoust. Soc. Am. 86, 1245–1253 (1989)] is used to model hydrophones embedded in viscoelastic layers of finite extent. The transfer function of the layer and/or the noise sensed by the hydrophone are computed for a given excitation of the outer surface or of the edges of the layer for different sizes. More complex multilayered structures are also examined. [Work partially supported by the Direction des Recherches Etudes et Techniques.]
88(1990); http://dx.doi.org/10.1121/1.2029121View Description Hide Description
The output (power) of an acoustic sensor, in response to the pressurefluctuations in a turbulent flow of fluid over its surface, is predicted by a certain integral over all possible wave numbers in both the flow direction and transverse to that direction. The integrand comprises four real and positive factors that describe (1) the pressure spectrum of the flow, (2) the filtering action of any material positioned between the sensors and the flow field, (3) the spatial averaging effect associated with the nonzero size of one of the sensors, and (4) the additional spatial averaging due to deploying not one, but an array of such sensors. Computation of the filtering action can be extremely time consuming for a multilayered medium. In addition, the rapid fluctuation of the array function with wave number gives rise to an integrand that cannot be efficiently integrated by standard numerical methods. This paper presents approximate techniques that allow for accurate and near real time evaluation of flownoise power spectral density. Some results are presented for a large array.
Response of an array of hydrophones embedded in an elastomer layer to a modified Corcos turbulent wall pressure spectrum88(1990); http://dx.doi.org/10.1121/1.2029122View Description Hide Description
A study was made of a theoretical model for evaluating turbulent boundary layerpressurefluctuations received by an array of rectangular hydrophones embedded in a layer of elastomer. The theoretical model considered in the present study is a plane elastomer layer backed with an infinite plate of finite thickness; the other side of the layer is exposed to turbulent flow. The transmitted flownoise was calculated with an empirically modified model of the Corcos turbulent wall pressure spectrum and was characterized by the frequency spectral density. The results presented here are numerical calculations of flow noise reductions, which are given relative to the noise levels computed for the flush‐mounted point hydrophone. Effects of the elastomer layer thickness, the backing plate thickness, the hydrophone dimensions, and the array dimensions on noise reduction are provided.