Multiple Hydrophones Research Articles

Cooperative-hydrophone-based underwater soundscape monitoring in a bustling harbor enhanced by coral ecosystem

Anthropogenic underwater noise is an emerging pollutant, urging a focus on studying the spatial and temporal dynamics of underwater soundscapes for effective environmental protection. This study centers on a bustling harbor enriched by coral ecosystems, employing 5–8 spatially distributed bottom-mounted passive acoustic monitoring (PAM) nodes strategically placed at locations of interest with diverse natural conditions and ship traffic intensities. Focused on the intricate interplay between anthropogenic and natural contributors, including shipping traffic and coral ecosystems, one-third octave band sound pressure levels SPLs were calculated and analyzed at different frequencies, revealing significant variations in levels across different times and locations. Diurnal variations in ambient noise levels highlight the complex relationship between human activities and marine ecosystems. Higher and more variable SPLs, primarily related to shipping traffic, were recorded during the daytime, while lower SPLs, accompanied by increased fish sounds, were observed at night. The collaborative deployment of multiple hydrophones enables an estimation of noise source direction, and multi-hydrophone based principal component analysis (PCA) for dimensionality reduction and subsequent clustering methods were used for noise source classification. This study significantly contributes to the field of underwater noise monitoring, offering a holistic perspective on busy harbor environments.

Bayesian detection and tracking of odontocetes in 3-D from their echolocation clicks.

Localization and tracking of marine animals can reveal key insights into their behaviors underwater that would otherwise remain unexplored. A promising nonintrusive approach to obtaining location information of marine animals is to process their bioacoustic signals, which are passively recorded using multiple hydrophones. In this paper, a data processing chain that automatically detects and tracks multiple odontocetes (toothed whales) in three dimensions (3-D) from their echolocation clicks recorded with volumetric hydrophone arrays is proposed. First, the time-difference-of-arrival (TDOA) measurements are extracted with a generalized cross-correlation that whitens the received acoustic signals based on the instrument noise statistics. Subsequently, odontocetes are tracked in the TDOA domain using a graph-based multi-target tracking (MTT) method to reject false TDOA measurements and close gaps of missed detections. The resulting TDOA estimates are then used by another graph-based MTT stage that estimates odontocete tracks in 3-D. The tracking capability of the proposed data processing chain is demonstrated on real acoustic data provided by two volumetric hydrophone arrays that recorded echolocation clicks from Cuvier's beaked whales (Ziphius cavirostris). Simulation results show that the presented MTT method using 3-D can outperform an existing approach that relies on manual annotation.

Source separation and arrival angle estimation using a single acoustic vector sensor

Extracting individual time signatures from two angularly-separated sources overlapping in time and bandwidth typically requires coherent array processing on multiple hydrophones, a process whose performance is degraded by limited aperture and spatial aliasing. Vector sensors, which measure both acoustic pressure and particle velocity, can measure an acoustic field’s directionality in a single, compact sensor package. Standard beamforming techniques can be applied to the outputs of a vector sensor to suppress one source’s signal, but its efficacy is limited by the angular insensitivity of the resulting cardioid beam pattern. Here we use fundamental physical principles to demonstrate that measurements on a 2-D vector sensor can completely determine the azimuths, amplitudes, and relative phases of two horizontal plane waves of the same frequency, achieving source separation and direction-of-arrival estimates for both. Interference between the waves creates a reactive intensity whose orientation is perpendicular to the bisector of the two wavenumber vectors. This observation leads to closed-form inversion formulas. Data examples of the technique are illustrated using overlapping humpback whale songs off Hawaii, and on weak airgun and bowhead whale signals embedded within intense directional noise from a nearby drilling site in the Beaufort Sea. [Work sponsored by ONR TFO.]

Comparison of imaging techniques for the analysis of bubble distribution in ship wakes

Ship wakes, which are inevitably caused by ship movement, are a subject of great interest for military purposes. In this study, we created an underwater acoustic signal measuring device to characterize the distribution of wakes from a ship maneuvering in a real ocean environment. Two sets of multi-array sonar systems consisting of multiple projectors and multiple hydrophones were fabricated to acquire underwater acoustic signals. The systems were then placed vertically at a certain distance from each other in the sea. Afterward, we stably acquired the transmitted and scattered signals from the bubble wake generated by an actual ship. The distribution of bubbles in the ship's wake was then imaged by applying the RTM (reverse-time migration) and WI (waveform inversion) numerical analysis techniques to image the bubble distribution in the ship wake. Finally, the results obtained using the two aforementioned techniques were analyzed and compared.

Effect of oceanography on broadband acoustic wave propagation in a range dependent shallow water environment

To investigate the effects of oceanographic variability on broadband acoustic wave propagation in shallow water environments, two experiments were carried out on the New England continental shelf region in 2017 and 2022, respectively. Towed sources were used to transmit chirp waveforms over the frequency ranges of 0.5—1.25 kHz and 1.5 – 4.0 kHz on the same acoustic tracks. The broadband acoustic transmissions were received by vertical acoustic arrays with multiple hydrophones. In 2017, the sound speed in the water column was isovelocity while in 2022 the presence of a duct at the depth of 20 m and upward refracting conditions reduced the interaction of acoustic transmissions with the seafloor compared to the experiment in 2017. Experimental observations show the importance of the water column on bottom interacting broadband waveforms causing challenges for the inversion algorithms. In this paper, we present acoustic and oceanographic results showing how the variability of sound speed could change the characteristics of the broadband transmission. Modeled results versus experimental data are discussed. [Work supported by ONR.]

HW/SW Platform for Measurement and Evaluation of Ultrasonic Underwater Communications.

The purpose of this work is to present a flexible system that supports the study of wideband underwater acoustic communications (UAC). It has been developed both to measure channels and to test transmission techniques under realistic conditions in the ultrasonic band. This platform consists of a hardware (HW) part that includes multiple hydrophones, projectors, analog front-ends, acquisition boards, and computers, and a software (SW) part for the generation, reception, and management of acoustic sounding signals and noise. UAC channels are among the most hostile ones and exhibit an important attenuation and distortion, essentially due to both multipath propagation, which results in a very long channel impulse response, and time-varying behavior, which produces a notable Doppler spread. To cope with this challenging medium, sophisticated transmission techniques must be employed. In this sense, adequate signal processing algorithms have been designed aiming not only at the analysis and characterization of underwater communication channels but also at the evaluation of diverse modulation, detection, and coding schemes, from Orthogonal Frequency Division Multiplexing (OFDM) to single-carrier digital modulations with a single-input multiple-output (SIMO) configuration that takes advantage of diversity techniques. Wideband sounding signals, to be injected into the sea from the transmitter side, are created with patterns that allow multiple tests on a batch. With offline processing of the captured data at the receiver side, different trials can be carried out in a very flexible manner. The different aspects of the platform are described in detail: the HW equipment used, the SW interface to control acquisition boards, and the signal processing algorithms to estimate the UAC channel response. The platform allows the analysis and design of new proposals for underwater communications systems that improve the performance of the current ones.

Multihydrophone Fusion Network for Modulation Recognition.

Deep learning (DL)-based modulation recognition methods of underwater acoustic communication signals are mostly applied to a single hydrophone reception scenario. In this paper, we propose a novel end-to-end multihydrophone fusion network (MHFNet) for multisensory reception scenarios. MHFNet consists of a feature extraction module and a fusion module. The feature extraction module extracts the features of the signals received by the multiple hydrophones. Then, through the neural network, the fusion module fuses and classifies the features of the multiple signals. MHFNet takes full advantage of neural networks and multihydrophone reception to effectively fuse signal features for realizing improved modulation recognition performance. Experimental results on simulation and practical data show that MHFNet is superior to other fusion methods. The classification accuracy is improved by about 16%.

The effects of array design on acoustic data collected during marine seismic reflection surveys

Marine seismic reflection surveys provide an abundance of acoustic data that are potentially useful for an array of analyses within and in addition to geoacoustic studies. A single cruise typically produces thousands to tens of thousands of acoustic events, with hundreds of hydrophones recording each event over towed arrays that can span up to 15 km or more. However, the structure of the airgun source arrays and the receiver arrays is typically not obvious from the data alone and the pre-processing of the data may yield misleading results if not properly accounted for. Generally, the acoustic source consists of an array of airguns configured such that the acoustic energy is focused towards the ocean floor, significantly impacting the relative intensity of the direct and reflected paths. Additionally, each array channel output typically consists of multiple hydrophone outputs, which are not individually available, averaged to generate a single channel output. Understanding the impact of these constructions is crucial for many analyses, such as mitigation of airgun pulse impacts on marine mammals. Here, we quantify the impact that these factors have on the output data and provide an analysis of this influence for an experimental case. [Work supported by ONR.]

Multiple Hydrophone Arrays based Underwater Localization with Matching Field Processing

Matched field processing technology (MFP) is a general passive localization method for underwater sound source due to its advantages in ultra-long distance positioning. In this paper, assume the total number of hydrophones remains unchanged, a single hydrophone array is divided into multiple hydrophone sub-arrays for independent positioning, and the positioning results of sub-arrays are fused to reduce the impact of noise and improve the robustness of the positioning system. Based on the traditional Bartlett processor, we derive the formula for average positioning error which varies with signal to noise ratio (SNR) and the number of hydrophones. The formula is used to decide the optimal structure of sub-arrays, i.e., the number of sub-arrays and the number of hydrophones in each sub-array. Experiments and simulations proves that multiple sub-arrays can improve the positioning accuracy compared with the single hydrophone array in the noisy environment. The average positioning errors produced by the experiments are consistent with the numerical ones based on the theoretical analysis.

Behavior related vocalizations of the Florida manatee (Trichechus manatus latirostris)

AbstractFlorida manatees (Trichechus manatus) produce five broadly defined call types (squeaks, squeals, high squeaks, chirps, squeak‐squeals) but their use in social and nonsocial settings is unclear. The goal of this study was to investigate whether call categories and structure of manatee vocalizations varied with behavior. Multiple hydrophones were used to record vocalizations in four different environments and broad behavioral states. Vocalizations recorded from resting, cavorting, stressed, or feeding wild animals were subjected to mixed linear effects models to test whether vocalizations produced varied with behavior and calf presence. Measures of duration, entropy, and frequency modulation were extracted from vocalizations to investigate if structural parameters differ among behaviors. Although all five call categories were recorded, results suggest manatees vocalize using primarily three call types and vary the structure of the call based on behavior. High squeaks were correlated with calf presence. High entropy squeals were proportionally higher during cavorting suggesting they may be related to a heightened state of arousal. Squeaks were the dominant call type produced and were longer in duration and higher in frequency modulation when animals were stressed. This research provides a foundation for comparative studies on vocal behavior for the Florida manatee as well as studies on related species.

Acoustic differentiation and classification of wild belugas and narwhals using echolocation clicks

Belugas (Delphinapterus leucas) and narwhals (Monodon monoceros) are highly social Arctic toothed whales with large vocal repertoires and similar acoustic profiles. Passive Acoustic Monitoring (PAM) that uses multiple hydrophones over large spatiotemporal scales has been a primary method to study their populations, particularly in response to rapid climate change and increasing underwater noise. This study marks the first acoustic comparison between wild belugas and narwhals from the same location and reveals that they can be acoustically differentiated and classified solely by echolocation clicks. Acoustic recordings were made in the pack ice of Baffin Bay, West Greenland, during 2013. Multivariate analyses and Random Forests classification models were applied to eighty-one single-species acoustic events comprised of numerous echolocation clicks. Results demonstrate a significant difference between species’ acoustic parameters where beluga echolocation was distinguished by higher frequency content, evidenced by higher peak frequencies, center frequencies, and frequency minimums and maximums. Spectral peaks, troughs, and center frequencies for beluga clicks were generally > 60 kHz and narwhal clicks < 60 kHz with overlap between 40–60 kHz. Classification model predictive performance was strong with an overall correct classification rate of 97.5% for the best model. The most important predictors for species assignment were defined by peaks and notches in frequency spectra. Our results provide strong support for the use of echolocation in PAM efforts to differentiate belugas and narwhals acoustically.

A Multi-Feature Compression and Fusion Strategy of Vertical Self-Contained Hydrophone Array

Applying passive sonar to classify underwater acoustic targets at different depths is a challenging task. Although the self-contained hydrophone array can ensure the normal operation of most units in various environments, it is arduous to achieve precise time synchronization between each hydrophone, which results in difficulties in data fusion between hydrophones. For a vertical sonar array composed of self-contained units, a deep learning-based data compression and multihydrophone fusion (DCMF) model is proposed to quickly extract acoustic propagation interference features, which are used for underwater acoustic target classification. Unlike the frequency-range domain striation features acquired by long-term accumulation, this paper exploits the depth difference between multiple hydrophones to obtain the frequency-depth domain joint striation features in a short time. The proposed DCMF conducts efficient feature compression and fusion via parallel stacked sparse autoencoders and a multi-input fusion network. The experimental results illustrate that the compressed features have strong robustness, a low mean square error with the simulation results, and shorter signal length requirements, which improves the classification efficiency and real-time performance of DCMF. In the case of the experimental dataset, DCMF is compared with several state-of-the-art multiscale fusion models, and the experiments indicate that DCMF has the best performance and smallest computational complexity.

Acoustic differentiation and classification of wild belugas and narwhals using echolocation clicks

Belugas (Delphinapterus leucas) and narwhals (Monodon monoceros) are highly social toothed Arctic cetaceans with large vocal repertoires and similar acoustic profiles. Passive Acoustic Monitoring (PAM) that uses multiple hydrophones over large spatiotemporal scales has been the primary method to study their populations, particularly in response to rapid climate change and increasing underwater noise. This study marks the first acoustic comparison between wild belugas and narwhals from the same location and reveals that they can be acoustically differentiated and classified solely by echolocation clicks. Acoustic recordings were made in the pack ice of Baffin Bay, West Greenland during 2013. Multivariate analyses and Random Forests classification models were applied to eighty-one single-species acoustic events comprised of numerous echolocation clicks. Results demonstrate a significant difference between species’ acoustic parameters where beluga echolocation was distinguished by higher frequency content. Classification model predictive performance was strong with an overall correct classification rate of 97.5% for the best model. The most important predictors for species assignment were defined by peaks and notches in frequency spectra. Our results provide strong support for the use of echolocation in PAM efforts to differentiate belugas and narwhals.

Tom Norris’s contributions to the acoustic density estimation of minke whales near Kauai, Hawaii

Tom Norris was involved in many collaborative efforts involving passive acoustic monitoring to study marine mammals. In the spring of 2010, he and his team conducted 1520 km of acoustic/visual line-transect surveys over 13 days of effort for minke whales (Balaenoptera acutorostrata) using the R/V Dariabar towing multiple hydrophones. The surveys were performed at the U.S. Navy Pacific Missile Range Facility’s offshore underwater range located north-west of Kauai, Hawaii. Minke whales have low observables and are rarely sighted in the Hawaii area which limits the available knowledge on their ecology. However they produce the distinctive boing vocalizations in the winter and spring seasons in the area, presumably for breeding purposes, which makes PAM methods attractive. Simultaneous with the acoustic line-transect surveys, the PMRF hydrophone data were also collected to estimate the boing vocalization density in a collaborative effort. Combining the minke whale density from the acoustic line-transect survey with the minke whale boing vocalization density provided an estimate for the call rate for boing calling minke whales which was a major accomplishment at the time.

Cross-correlation analysis of multiple fibre optic hydrophones for water pipeline leakage detection

Leakage detection plays a significant role in water pipeline since leakage problems severely affect urban water supply safety. The fibre optic hydrophone derives from fibre optic hydroacoustic sensing technology and exhibits superior performance to traditional electric hydrophones. To investigate leakage detection performance of the proposed fibre optic hydrophone system, tests were conducted on overhead and buried pipelines, respectively. Two types of spherical valves were adopted for simulating leakage with different sizes. Effects of surrounding medium on leakage acoustic wave and positioning accuracy were studied. The signal was acquired by hydrophone analyser and analysed by a self-developed program. A revised cross-correlation analysis of multi-signals was proposed and compared with that of two signals; it presented a better performance in time delay estimation. With leakage, overhead pipelines showed larger amplitudes than that of buried pipelines and signal amplitude increased with the opening degree of valves. The results reveal that the surrounding medium of the pipeline affected the leakage wave velocity and could attenuate the wave. The cross-correlation analysis of multiple hydrophones could increase the accuracy of leakage positioning. The proposed leakage detection system was feasible to detect leakage of both overhead and buried pipelines.

Source spatial averaging for the sensitivity measurement of multiple hydrophones in a tank

In this paper, we propose a novel calibration method for the batch measurement of hydrophone sensitivity in a tank. This calibration method is based on the source spatial averaging method of constructing a ‘diffused sound field’ in a small-volume tank. Source spatial averaging is implemented as follows: the transducer radiates steady-state signals whilst moving irregularly and slowly at a certain space, and average pressure is calculated equally at each position by average-processing the energy density in the reverberation control region. For demonstration, an experiment was conducted to calibrate six hydrophones in a glass tank with a dimension of 1.5 m × 0.9 m × 0.6 m and a frequency band range of 5–20 kHz. The measured readings were consistent with the factory-product identification. Error value was less than ± 0.5 dB, and standard uncertainty was less than 0.95 dB. Results demonstrate the accuracy and viability of the proposed batch calibration method based on source spatial averaging. The method can effectively and simultaneously calibrate multiple hydrophones in limited tank space and improve working efficiency.

Variability of radiated underwater noise measurements for a small research vessel in shallow water

The increased interest in the potential environmental impact of noise from shipping is resulting in the development of measurement methods for determining the radiated noise level and equivalent monopole source level of vessels. It is important to understand the random and systematic uncertainties associated with such techniques. The EU SONIC project provided an opportunity to make repeated measurements on the University of Newcastle research vessel, the Princess Royal, in a shallow water environment (100 m deep) over a number of days. Two multiple hydrophone arrays operated by two of the project partners, CETENA and the University of Southampton, were deployed from the same moored support vessel. This data is reviewed to illustrate the variation associated with such measurements and the impact of using multiple hydrophones and multiple measurement runs on the uncertainty in the radiated noise level and the calculated source level. The results are also compared with a later measurements of the same vessel in a different shallow water environment (20 m deep) using bottom moored hydrophones.

Passive acoustic monitoring shows no effect of anthropogenic noise on acoustic communication in the invasive round goby (Neogobius melanostomus)

Abstract Passive acoustic monitoring (PAM) can be a powerful tool to survey populations of soniferous species in natural settings. Using PAM allows monitoring of behaviour and activity unobtrusively to get a more accurate assessment of behavioural responses than traditional techniques. The use of PAM can also be beneficial for behaviourally cryptic species or other taxa that might be missed by more traditional sampling methods. To quantify seasonality of reproduction in an aquatic invasive species and to assess possible interactions between both abiotic and biotic noise and acoustic ecology, we deployed multiple hydrophones to fully characterise the acoustic behaviour of round goby (Neogobius melanostomus), a highly invasive species. We also aimed to quantify levels of anthropogenic noise around the goby communities. We tested correlations of round goby calling rates with overall noise levels and the presence of boat traffic as direct tests of the hypothesised role of noise on natural acoustic behaviour of fish. Round goby showed a clear diel patterning of calling behaviour, with the highest activity during the night and ceasing at midday; supporting the importance of acoustic—as opposed to visual—signalling for mate attraction. Calls also allowed us to pinpoint spawning areas for possible remediation. We saw no correlation between measures of background noise and calling behaviour or presence of boats and calling, suggesting that increased levels of noise have no effect on the natural calling behaviour of this species. Passive acoustic monitoring of round goby calling behaviour is a promising technique for those interested in remediation of this highly invasive species, as it can be difficult to quantify population levels by more conventional means. The lack of noise effects we see suggests that noise from recreational boats does not disrupt signalling in this species with limited hearing range, although more investigation is needed to ascertain whether noise could be a stressor in other contexts.

Deep learning for ethoacoustical mapping: Application to a single Cachalot long term recording on joint observatories in Vancouver Island

During February and March, 2018, a lone sperm whale known as Yukusam was recorded first by Orcalab in Johnstone Strait and subsequently on multiple hydrophones within the Salish Sea [1]. We learn and denoise these multichannel clicks trains with AutoEncoders Convolutional Neural Net (CNN). Then, we build a map of the echolocations to elucidate variations in the acoustic behavior of this unique animal over time, in different environments and distinct levels of boat noise. If CNN approximates an optimal kernel decomposition, it requires large amounts of data. Via spline functionals we offer analytics kernels with learnable coefficients do reduce it. We [1-3] identify the analytical mother wavelet to represent the input signal to directly learn the wavelet support from scratch by gradient descend on the parameters of cubic splines [2]. Supplemental material http://sabiod.org/yukusam [1] Balestriero, Roger, Glotin, Baraniuk, Semi-Supervised Learning via New Deep Network Inversion, arXiv preprint arXiv:1711.04313, 2017 [2] Balestriero, Cosentino, Glotin, Baraniuk, WaveletNet : Spline Filters for End-to-End Deep Learning, Int. Conf. on MachineLearning, ICML, Stockholm, http://sabiod.org/bib, 2018 [3] Spong P., Symonds H., et al., Joint Observatories Following a Single male Cachalot during 12 weeks—The Yukusam story, ASA 2018.

Using machine learning in ocean noise analysis during marine seismic reflection surveys

Marine seismic reflection surveys use repetitive broadband sound signals to image the structure of the seafloor. High energy sound signals generated by airguns are recorded by single or multiple long horizontal hydrophone arrays towed behind the vessel. Marine seismic reflection surveys raise concern over their effects on marine animals. In particular, there is concern about (1) the impact of the airgun sound power level on marine mammals and (2) the increase in the background noise level caused by the reverberation between shots which may mask marine mammals’ vocalizations and degrade the performance of the passive acoustic monitoring systems. Quantifying these effects is a challenging task due to the complexity of local geology, seafloor topography, and uncertainty in water properties. In this study, a machine learning approach is used to predict the airgun sound power level and the reverberation level off the ocean floor. The data utilized in this study are from the COAST (Cascadia Open-Access Seismic Transects) seismic reflection survey conducted with the R/V Marcus Langseth in July 2012. This experiment spanned a wide range of water depths from the continental shelf to deep water. Data was recorded by an 8-km horizontal array with 636 hydrophones sampled at 500 Hz.