Search

The Global Navigation Satellite System—Acoustic ranging combination technique (GNSS-A) is a seafloor geodetic observation technique that achieves an accuracy of centimetres by combining high-rate GNSS data with acoustic ranging. The technique determines the seafloor position by acoustic ranging between the sea surface and multiple seafloor stations, using GNSS data from the sea surface station. Here, the gradient state of the underwater sound speed structure (SSS) is a significant source of error. The open-source software GARPOS can reduce the effect from underwater gradient structures by retrieving the underwater disturbance as a parameter projected onto the sea surface and seafloor. To evaluate the effects of underwater disturbances, a quantitative comparison of the model parameters is necessary. In this study, we developed a representation method to evaluate features of the ocean field. Here, the expression method was described in the order of a formulation and an interpretation in the case of a 2-D cross-section and extension to the case of an actual 3-D field. This method makes it possible to evaluate SSS states in the GNSS-A observations. As an example, we showed the correlation between the anomaly of the expressed ocean state and the anomaly of the seafloor station position, showing that this expression method is an effective index for correcting bias errors. Additionally, we used the data from sites located in the Kuroshio area, a strong current near Japan, to show that the ocean state can be quantitatively interpreted using this expression method. [ABSTRACT FROM AUTHOR]

The probability distribution of three-dimensional sound speed fields (3D SSFs) in an ocean region encapsulates vital information about their variations, serving as valuable data-driven priors for SSF inversion tasks. However, learning such a distribution is challenging due to the high dimensionality and complexity of 3D SSFs. To tackle this challenge, we propose employing the diffusion model, a cutting-edge deep generative model that has showcased remarkable performance in diverse domains, including image and audio processing. Nonetheless, applying this approach to 3D ocean SSFs encounters two primary hurdles. First, the lack of publicly available well-crafted 3D SSF datasets impedes training and evaluation. Second, 3D SSF data consist of multiple 2D layers with varying variances, which can lead to uneven denoising during the reverse process. To surmount these obstacles, we introduce a novel 3D SSF dataset called 3DSSF, specifically designed for training and evaluating deep generative models. In addition, we devise a high-capacity neural architecture for the diffusion model to effectively handle variations in 3D sound speeds. Furthermore, we employ state-of-the-art continuous-time-based optimization method and predictor-corrector scheme for high-performance training and sampling. Notably, this paper presents the first evaluation of the diffusion model's effectiveness in generating 3D SSF data. Numerical experiments validate the proposed method's strong ability to learn the underlying data distribution of 3D SSFs, and highlight its effectiveness in assisting SSF inversion tasks and subsequently characterizing the transmission loss of underwater acoustics. [ABSTRACT FROM AUTHOR]

The conversion from seismic to ocean-acoustic waves occurs in different places on the bottom of the ocean, often hundreds to thousands of kilometers away from the epicenter. Here, we investigate this conversion process by studying 15 large-magnitude earthquakes that occurred between 2014 and 2022 along the Kermadec Arc in the southwestern Pacific Ocean. To pinpoint the location where seismic-to-acoustic conversion takes places, we analyze hydroacoustic signals recorded by a hydrophone triplet station of the International Monitoring System in the Juan Fernández archipelago. Results from direction-of-arrival and travel-time calculations indicate that the location of the conversion zone largely matches segments of the Louisville Seamount Chain, its lateral extent ranging from approximately 300 to 1800 km, and its location depending on the geometry between earthquake epicenter and the seamounts. [ABSTRACT FROM AUTHOR]

Sound speed is a critical parameter in ocean acoustic studies, as it determines the propagation and interpretation of recorded sounds. The potential for exploiting oceanic vessel noise as a sound source of opportunity to estimate ocean sound speed profile is investigated. A deep learning-based inversion scheme, relying upon the underwater radiated noise of moving vessels measured by a single hydrophone, is proposed. The dataset used for this study consists of Automatic Identification System data and acoustic recordings of maritime vessels transiting through the Santa Barbara Channel between January 2015 and December 2017. The acoustic recordings and vessel descriptors are used as predictors for regressing sound speed for each meter in the top 200 m of the water column, where sound speeds are most variable. Multiple (typically ranging between 4 and 10) transits were recorded each day; therefore, this dataset provides an opportunity to investigate whether multiple acoustic observations can be leveraged together to improve inversion estimates. The proposed single-transit and multi-transit models resulted in depth-averaged root-mean-square errors of 1.79 and 1.55 m/s, respectively, compared to the seasonal average predictions of 2.80 m/s. [ABSTRACT FROM AUTHOR]

Simulating the acoustic field excited by pulse sound sources holds significant practical value in computational ocean acoustics. Two primary methods exist for modeling underwater acoustic propagation in the time domain: the Fourier synthesis technique based on frequency decomposition and the time-domain underwater acoustic propagation model (TD-UAPM). TD-UAPMs solve the wave equation in the time domain without requiring frequency decomposition, providing a more intuitive explanation of the physical process of sound energy propagation over time. However, time-stepping numerical methods can accumulate numerical errors, making it crucial to improve the algorithm's accuracy for TD-UAPMs. Herein, the time-domain wavenumber integration model SPARC was improved by replacing the second-order finite element method (FEM) with the high-order accuracy finite difference method (FDM). Furthermore, the matched interface and boundary (MIB) method was used to process the seabed more accurately. The improved model was validated using three classic underwater acoustic benchmarks. By comparing the acoustic solutions obtained using the FDM and the FEM, it is evident that the improved model requires fewer grid points while maintaining the same level of accuracy, leading to lower computational costs and faster processing compared to the original model. [ABSTRACT FROM AUTHOR]

When an airborne sound source is in rapid motion, the acoustic signal detected by the underwater sensor experiences a substantial Doppler shift. This shift is intricately linked to the motion parameters of the sound source. Analyzing the Doppler shift characteristics of received acoustic signals enables not only the estimation of target motion parameters but also the localization of the airborne sound source. Currently, the predominant methods for estimating parameters of uniformly moving targets are grounded in classical approaches. In this study, the application of the Doppler warping transform, traditionally applicable to sound sources in uniform linear motion, is extended to encompass a broader spectrum of sound source trajectories. Theoretical and experimental data validate the efficacy of this transform in linearizing the Doppler shift induced by a source in curved motion. Building upon this foundation, a methodology is proposed for locating airborne acoustic sources in curved motion from underwater. Sea experimental data corroborate the method's effectiveness in achieving underwater localization of a helicopter target during curved motion. [ABSTRACT FROM AUTHOR]

Multibeam imaging SONARs have been used for a range of measurement applications, such as measurements of fish lengths. This study aimed to quantify the accuracy of imaging SONAR systems, that varied in frequency and beam geometry, to measure the length of synthetic targets positioned perpendicularly.Blueprint Oculus imaging SONAR systems, with four different (centre) frequencies (750 kHz, 1.2 MHz, 2.1 MHz and 3 MHz), were used to measure the length of three targets of nominal lengths: 10 cm, 20 cm and at ranges between 1 and 15.5 m. The effect of beam geometry on measurement error was then examined using regression analysis.This study found that there was an overestimation of the actual length of the target for all measurements that was inversely proportional to the horizontal beamwidth. The measurement error can be reduced by normalising for beamwidth. However, the variation of measurements (i.e. the precision) was found to also increase with range, which was attributed to the increasing beam separation.It is important that the effect of beamwidth on the accuracy of target length measurements, by imaging SONARs is acknowledged in future studies. One approach to mitigating this problem, is to limit the range at which length measurements are made to a beamwidth that produces an estimated level of error that is acceptable for that study. [ABSTRACT FROM AUTHOR]

Side-scan sonar (SSS) is now a prevalent instrument for large-scale seafloor topography measurements, deployable on an autonomous underwater vehicle (AUV) to execute fully automated underwater acoustic scanning imaging along a predetermined trajectory. However, SSS images often suffer from speckle noise caused by mutual interference between echoes, and limited AUV computational resources further hinder noise suppression. Existing approaches for SSS image processing and speckle noise reduction rely heavily on complex network structures and fail to combine the benefits of deep learning and domain knowledge. To address the problem, RepDNet, a novel and effective despeckling convolutional neural network is proposed. RepDNet introduces two re-parameterized blocks: the Pixel Smoothing Block (PSB) and Edge Enhancement Block (EEB), preserving edge information while attenuating speckle noise. During training, PSB and EEB manifest as double-layered multi-branch structures, integrating first-order and secondorder derivatives and smoothing functions. During inference, the branches are re-parameterized into a 3 x 3 convolution, enabling efficient inference without sacrificing accuracy. RepDNet comprises three computational operations: 3 x 3 convolution, element-wise summation and Rectified Linear Unit activation. Evaluations on benchmark datasets, a real SSS dataset and Data collected at Lake Mulan aestablish RepDNet as a well-balanced network, meeting the AUV computational constraints in terms of performance and latency. [ABSTRACT FROM AUTHOR]

Predicting acoustic transmission loss in the SOFAR channel faces challenges, such as excessively complex algorithms and computationally intensive calculations in classical methods. To address these challenges, a deep learning-based underwater acoustic transmission loss prediction method is proposed. By properly training a U-net-type convolutional neural network, the method can provide an accurate mapping between ray trajectories and the transmission loss over the problem domain. Verifications are performed in a SOFAR channel with Munk's sound speed profile. The results suggest that the method has potential to be used as a fast predicting model without sacrificing accuracy. [ABSTRACT FROM AUTHOR]

We examine the dependence of the penetration depth and fractional surface area (e.g., whitecap coverage) of bubble plumes generated by breaking surface waves on various wind and wave parameters over a wide range of sea state conditions in the North Pacific Ocean, including storms with sustained winds up to 22 m s−1 and significant wave heights up to 10 m. Our observations include arrays of freely drifting SWIFT buoys together with shipboard systems, which enabled concurrent high‐resolution measurements of wind, waves, bubble plumes, and turbulence. We estimate bubble plume penetration depth from echograms extending to depths of more than 30 m in a surface‐following reference frame collected by downward‐looking echosounders integrated onboard the buoys. Our observations indicate that mean and maximum bubble plume penetration depths exceed 10 and 30 m beneath the surface during high winds, respectively, with plume residence times of many wave periods. They also establish strong correlations between bubble plume depths and wind speeds, spectral wave steepness, and whitecap coverage. Interestingly, we observe a robust linear correlation between plume depths, when scaled by the total significant wave height, and the inverse of wave age. However, scaled plume depths exhibit non‐monotonic variations with increasing wind speeds. Additionally, we explore the dependencies of the combined observations on various non‐dimensional predictors used for whitecap coverage estimation. This study provides the first field evidence of a direct relation between bubble plume penetration depth and whitecap coverage, suggesting that the volume of bubble plumes could be estimated by remote sensing. Plain Language Summary: Quantifying the statistics of bubble plumes generated during ocean surface wave breaking is essential to understanding the exchange between the ocean and the atmosphere. Bubble plumes also cause important variations in underwater acoustics and optics. Recent studies also suggest that the statistics of bubble plumes are skillful predictors for total energy loss during wave breaking, which is an essential quantity for accurate wave forecasting. In this study, we examine how these bubble plume statistics change with different wind and wave conditions, including during storms. We used echosounders on drifting buoys to detect the bubbles and estimate how deep they go in the ocean. We also used shipboard camera systems to measure the surface area of the bubble plumes. We successfully develop multiple empirical relationships that allow us to predict how bubble plume depth and surface area change as a function of simple wind and wave statistics. These statistics are readily available from existing forecast models or typical ocean buoys. Our findings reveal that bubble plume depth is correlated with its visible surface area. This intriguing correlation suggests that we might estimate the volume of bubble plumes simply by observing the ocean surface from above. Key Points: Bubble plumes generated during ocean surface wave breaking are observed with echosounders on drifting buoysBubble plume depths are well correlated with whitecap coverage, wind speed, and spectral wave steepnessBubble plumes persist for many wave periods and exceed the persistence of visible surface foam [ABSTRACT FROM AUTHOR]

To evaluate underwater sound, a simple non-invasive optical technique based on self-interference to detect low-frequency underwater acoustic signals was demonstrated. Clear self-interference fringes of a laser beam reflected from the surface capillary wave transformed from a low-frequency underwater acoustic signal by a cylinder were observed. This study was also aimed at developing an artificial neural network (ANN) model with input from an optical pattern. The relationship between the optical pattern and the frequency of underwater sound was established, and an optimal combination of hyper-parameters was obtained. By analyzing the fringe region and the fringe interval, the frequency of the underwater acoustic signal and its relative amplitude were measured. A model based on physical optics modified the fringe distribution function, and the theoretical fit with the modified function was in good agreement with experimental observation. The BP-ANN model exhibited good performance in establishing the relationship between the optical pattern and underwater sound. [ABSTRACT FROM AUTHOR]

A hydrophone is a core component of sonar and plays a significant role in underwater detection. High sensitivity and wide‐frequency responsiveness are crucial parameters for underwater mechanosensors. This study introduces an innovative wood‐based hydrophone for precise detection of water wave spectra and a wide range of underwater sounds. The wood hydrophone consists of a conductive wood electrode with deposited polypyrrole (PPy) microbranches (polypyrrole wood (PW)) and a cellulosic micropatterned anti‐swelling hydrogel (MPASH) serving as the dielectric layer. As pressure increases within the range of 6.4 Pa to 1.2 kPa, the densely distributed PPy microbranches within the wood fiber cell lumina contact with more tips of the self‐wrinkled micropatterns of MPASH. This interaction results in a four‐order‐of‐magnitude increase in capacitance, leading to an unprecedented sensitivity of 1153 kPa−1 toward microstresses. Due to the presence of multisized configurations in both PW and MPASH, the sensitivity of the wood hydrophone decreases gradually (268 kPa−1 up to 37 kPa). The receiving sensitivity of the wood hydrophone for acoustic waves is greater than ‐174 dB within the frequency range of 250–2300 Hz. This work presents a promising avenue for the development of underwater detection devices utilizing graded wood channels and micropatterned cellulosic hydrogels. [ABSTRACT FROM AUTHOR]

Harbor porpoises (Phocoena phocoena) are the only cetaceans routinely sighted in Hood Canal, a narrow fjord that comprises the western edge of Puget Sound, Washington, USA. Harbor porpoises are sensitive to anthropogenic sounds, including noise from recreational and commercial vessel traffic, and the United States Navy, which conducts military training and testing within Hood Canal that can include underwater sound sources. This study was funded as part of the Navy monitoring program to assess potential impacts of naval activities on cetaceans. We conducted vessel-based line-transect surveys for harbor porpoises in Hood Canal in 2022-2023 to derive seasonal estimates of abundance and density. We carried out surveys over 37 days and surveyed the entire canal twice per season totaling 2,176 km of on-effort track line. We recorded 809 oneffort harbor porpoise groups and 1,385 individuals. Seasonal abundance estimates were lowest in winter (308 animals, 95% CI = 189-503) and gradually increased through spring and summer to a peak of 1,336 animals (95% CI = 826-2,160) in fall. Overall porpoise density was highest in central Hood Canal, an area that includes a designated United States Navy training range, though porpoise sightings were notably absent in a 21-km² area adjacent to the naval submarine base within this otherwise high-density region. Though we collected only a single year of data, these results suggest that harbor porpoise abundance in Hood Canal increased significantly since it was last estimated (2013-2015). The notable seasonal fluctuation of harbor porpoise abundance suggests Hood Canal may host a larger percentage of the overall Washington Inland Waters stock during the fall season, raising important management considerations. [ABSTRACT FROM AUTHOR]

To further design and develop broadband and efficient underwater sound absorption structure, we propose a hybrid acoustic structure composed of a rubber matrix layer, cylindrical cavity, pentamode metamaterial (PM) layer, impedance matching layer, and steel backing plate. Thanks to the special properties of the PMs, the hybrid acoustic structure obtains excellent broadband sound absorption performance (SAP) in the frequency range of 500–10000 Hz than that of the traditional cavity acoustic structure. To this end, we study the design principle of PM structure and its metal–water characteristics and analyze the sound absorption mechanism at the working frequency. In addition, the research shows that the material and geometric parameters of the structure have a significant impact on the SAP. By the optimization of structural material parameters, the average absorption coefficient at 500–10000 Hz can further increase to above 0.781. Finally, we also demonstrate that the structure still has good SAP under a wide range of oblique incidence angles. The research in this paper opens up bright perspectives for the design of underwater anechoic coatings for multifunctional applications. [ABSTRACT FROM AUTHOR]

As the global effects of climate change become more known year by year, it becomes ever-more pertinent to examine the effects this may bring for every aspect of modern life we rely on. One topic of focus is that of multi-frequency sonar communication and navigational systems, which rely on well-established relationships relating to wave speed, signal intensity, and attenuation. We compiled data on oceanic temperature, acidity, and salinity in the Upper European Shelf, which includes the North Sea and Mediterranean Sea, from 2006 to 2072 using the CMIP5 future climate model in the RCP8.5 scenario. We calculate that the speed of sound in the northern European oceanic area will decrease by almost 18 m/s by 2072, with an average yearly decrease in sound speed by 0.37 m/s. The attenuation of sound through water will change year by year, calculated based on a higher-order polynomial regression dependent on the frequency of sonar used. The maximum operating ranges of active low-frequency, mid-frequency, and high-frequency sonar systems would theoretically change by +0.06%, -0.19%, and +0.71%, respectively per year, if no other factors are affected. Due to increased sound propagation, the ambient noise level of the ocean would also increase and have some counter-effect to the increased detection range however that increase in noise level was not quantitatively analyzed in this study. [ABSTRACT FROM AUTHOR]

An algorithm has been substantiated for passive determining the distance to sound sources in convergence zones (CZ), which are observed in most deep-sea areas of the World Ocean. The algorithm is based on the well-known pattern of formation of the ray structure of the acoustic field of the source signal at the input of the receiving array in CZ. The pattern is as follows: when a source enters a CZ by crossing its near boundary, the maximum energy of its signal arrives at the array along rays with negative grazing angles, and when the source enters a CZ by crossing its far boundary, along rays with positive grazing angles. The accuracy in determining distance using the proposed algorithm for detecting surface and underwater sources is assessed. [ABSTRACT FROM AUTHOR]

On the example of an underwater sound channel typical of the Philippine Sea [1–3], it was established by numerical simulation using mode theory that during experimental studies of the propagation of explosive signals by R.A. Vadov [1–3], manifestation of a weakly divergent beam in the spatiotemporal structure of the acoustic field was observed for the first time: at certain locations of the corresponding points in the oceanic waveguide, in addition to classical quadruples of pulses, additional acoustic signals with small time delays with respect to them. [ABSTRACT FROM AUTHOR]

The parabolic wave equation describes wave propagation in a preferable direction, which is usually horizontal in underwater acoustics and vertical in seismic applications. For dense receiver arrays (receiver spacing is less than signal wavelength), this equation can be used for propagating the recorded wavefield back into the medium for imaging sources and scattering objects. Similarly, for multiple source and receiver array acquisition, typical for seismic exploration and potentially beneficial for ocean acoustics, one can model data in one run using an extension of the parabolic equation—the pseudo-differential Double Square Root (DSR) equation. This extended equation allows for the modeling and imaging of multi-source data but operates in higher-dimensional space (source, receiver coordinates, and time), which makes its numerical computation time-consuming. In this paper, we apply a faster ray method for solving the DSR equation. We develop algorithms for both kinematic and dynamic ray tracing applicable to either data modeling or true-amplitude recovery. Our results can be used per se or as a basis for the future development of more elaborated asymptotic techniques that provide accurate and computationally feasible results. [ABSTRACT FROM AUTHOR]

Underwater localization is one of the key techniques for positioning, navigation, timing (PNT) services that could be widely applied in disaster warning, underwater rescues and resource exploration. One of the reasons why it is difficult to achieve accurate positioning for underwater targets is due to the influence of uneven distribution of underwater sound velocity. The current sound-line correction positioning method mainly aims at scenarios with known target depth. However, for nodes that are non-cooperative nodes or lack depth information, sound-line tracking strategies cannot work well due to non-unique positional solutions. To solve this problem, we propose an iterative ray tracing 3D underwater localization (IRTUL) method for stratification compensation. To demonstrate the feasibility of fast stratification compensation, we first derive the signal path as a function of initial |grazing angle, and then prove that the signal propagation time and horizontal propagation distance are monotonic functions of the initial grazing angle, which guarantees the fast achievement of ray tracing. Simulation results indicate that IRTUL has the most significant correction effect in the depth direction, and the average accuracy has been improved by about 3 m compared to a localization model with constant sound velocity. [ABSTRACT FROM AUTHOR]

Degradation of polymer composites is a significant problem in many engineering aspects. Due to the interaction of various degradation factors during the exploitation of composites, a synergistic effect of destruction is observed. The article describes the phenomena occurring in glass fiber reinforced polyester laminates under the influence of ultraviolet radiation (UV) in an aquatic environment. The laminates were exposed to UV-A, UV-B and UV-C radiation for 1000 h in free-air and underwater conditions. During the test, the materials were immersed at stable depth of 1 mm and 10 mm, respectively. The three-point bending tests performed on the samples after being exposed to UV showed an increase in the flexural strength of the composites. Simultaneously, degradation of the outer surface layer was observed. The degradation removed the thin resin film from the surface which resulted in a direct exposure of the reinforcing fibers to the environment. The transformations taking place in the deeper layers of the composite increased the mechanical strength due to the additional cross-linking reactions excited by the energy arising from the radiation. Moreover, the formation of polymer structures from free styrene remaining after the technological process and the occurrence of free radical reactions as a result of the cage effect was also observed. [ABSTRACT FROM AUTHOR]

Since the industrial revolution the ocean has become noisier. The global increase in shipping is one of the main contributors to this. In some regions, shipping contributed to an increase in ambient noise of several decibels, especially at low frequencies (10 to 100 Hz). Such an increase can have a substantial negative impact on fish, invertebrates, marine mammals and birds interfering with key life functions (e.g. foraging, mating, resting, etc.). Consequently, engineers are investigating ways to reduce the noise emitted by vessels when designing new ships. At the same time, since the industrial revolution (starting around 1760) greenhouse gas emissions have increased the atmospheric carbon dioxide fraction x(CO2) by more than 100 mmol mol-1. The ocean uptake of approximately one third of the emitted CO2 decreased the average global surface ocean pH from 8.21 to 8.10. This decrease is modifying sound propagation, especially sound absorption at the frequencies affected by shipping noise lower than 10 kHz, making the future ocean potentially noisier. There are also other climate change effects that may influence sound propagation. Sea surface warming might alter the depth of the deep sound speed channel, ice melting could locally decrease salinity and more frequent storms and higher wind speed alter the depth of the thermocline. In particular, modification of the sound speed profile can lead to the appearance of new ducts making specific depths noisier. In addition, ice melting and the increase in seawater temperature will open new shipping routes at the poles increasing anthropogenic noise in these regions. This review aims to discuss parameters that might change in the coming decades, focusing on the contribution of shipping, climate change and economic and technical developments to the future underwater soundscape in the ocean. Examples are given, contrasting the open ocean and the shallow seas. Apart from the changes in sound propagation, this review will also discuss the effects of water quality on ship-radiated noise with a focus on propeller cavitation noise. [ABSTRACT FROM AUTHOR]

Freshwater fishes exhibit a wide range of auditory adaptations and capabilities, which are assumed to help them navigate their environment, avoid predators, and find potential mates. Yet, we know very little about how freshwater environments sound to fish, or how fish with different auditory adaptations respond to different soundscapes. We first compiled data on fish hearing acuity and adaptations and provided a portrait of how anthropogenic sounds compare to natural sounds in different freshwater soundscapes. We then conducted a sound‐enrichment field experiment at Lake Saint Pierre, a large fluvial lake in Canada, to evaluate the effect of motorboat sound exposure on the fish community by looking at the extent to which changes in species abundances were linked to auditory adaptations. Data compilation showed that the hearing acuity of most species overlaps with a wide range of ambient and anthropogenic underwater sounds while the field experiment showed that species with more specialized auditory structures were captured less often in sound‐enriched traps, indicating avoidance behavior. Our findings highlight the importance of considering species' sensorial adaptations when evaluating the community‐scale effects of anthropogenic sounds on the fish community, especially at low levels of anthropogenic activity. [ABSTRACT FROM AUTHOR]

Underwater sound speed is one of the most significant factors that affects high-accuracy underwater acoustic positioning and navigation. Due to its complex temporal variation, the forecasting of the underwater sound speed field (SSF) becomes a challenging task. Taking advantage of machine learning methods, we propose a new method for SSF forecasting based on the least square support vector machine (LSSVM) and a multi-parameter model, aiming to enhance the forecasting accuracy of underwater SSF with hourly resolution. We first use a matching extension method to standardize profile data and train the LSSVM with the parameters of observation time, temperature, salinity, and depth. We then employ radial basis function kernels to construct the forecasting model of SSF. We validate the feasibility and effectiveness of the LSSVM model by comparing it with the polynomial fitting (PF) and back propagation neural network (BPNN) methods, using hourly data obtained from the measured data and open data. The results show that the means of the root mean square for the LSSVM based on the observation time parameter and the LSSVM based on the multi-parameter model achieve 0.51 m/s and 0.45 m/s, respectively, presenting a significant improvement compared with the PF (0.82 m/s) and BPNN (0.76 m/s) methods. [ABSTRACT FROM AUTHOR]

The underwater channel is bilateral, heterogeneous, uncertain, and exhibits multipath transmission, sound line curvature, etc. These properties complicate the structure of the received pulse, causing great challenges in direct signal identification for ranging purposes and impacts on back-end data processing, even accurate acoustic positioning. Machine learning (ML) combined with underwater acoustics has emerged as a prominent area of research in recent years. From a statistical perspective, ML can be viewed as an optimization strategy. Nevertheless, the existing ML-based direct-signal discrimination approaches rely on independent assessment, utilizing a single sensor (beacon or buoy), which is still insufficient for adapting to the complex underwater environment. Thus, discrimination accuracy decreases. To address the above issues, an accurate CW direct signal detection approach is performed using the decision tree algorithm, which belongs to ML. Initially, the pulse parameter characteristics in the underwater multipath channel are investigated and the parameter models are built. Then, based on multi-sensor localization performance feedback, fusion characteristics for diverse pulse are created. Next, the pulse parameter characteristics are preprocessed to mitigate the impact of varying magnitudes and units of magnitude on data processing. Then, the decision tree is built to obtain the desired output results and realize accurate recognition of the ranging direct signals. Finally, the feasibility and reliability of this paper's method are verified by computer simulation and field testing. [ABSTRACT FROM AUTHOR]

Obtaining accurate sound speed profiles (SSPs) in near-real-time is of great significance for ocean exploration, underwater communication and improving the performance of sonar systems. In response to the problem that traditional sound speed estimation methods cannot obtain real-time sound speed distribution or rely too much on sonar observation data, we propose an SSP estimation method based on a convolutional neural network with reduced fully connected layers (RFC-CNN) in this paper. This method utilizes neural networks to extract the complex nonlinear features of various types of data. With the help of the historical SSPs and shallow seawater sound speed and temperature data obtained by expendable conductivity–temperature–depth probes (XCTDs), a more accurate estimation of the regional sound speed distribution can be realized quickly. This approach can save the observation cost and significantly improve the real-time performance of SSP estimation. [ABSTRACT FROM AUTHOR]

Acoustic propagation through a random distribution of 1 m ice cubes, from 100 to 1000 Hz, was simulated in a 3D finite element model. The effective sound speed and attenuation as functions of frequency were calculated from the simulated signals. Attempts were made to fit a number of models to the wave speed and attenuation, including single scattering, lossy water, and Biot approximations. An extended Biot model, developed for acoustic propagation in granular seabed sediments, was able to fit the simulation up to 300 Hz. Beyond this frequency, the simulation shows that multiple scattering dominates. [ABSTRACT FROM AUTHOR]

When a vibrating structure is located in a half-space bounded by a reflecting surface, the measured sound pressures are polluted by the reflection, so that they cannot show the actual sound radiation from the structure. The half-space spherical wave function expansion-based near-field acoustical holography approach is proposed to reconstruct the direct radiation from the sound source using the polluted sound pressures, which has been validated by numerical simulations. This paper deals with the experimental validation of this approach applied in an underwater halfspace bounded by a water-air interface. The sound pressures measured in the half-space are used as the input data. The reconstructed sound pressures obtained are compared with benchmarks measured in a free space. The results indicate that the influences of the water-air interface reflection on the measured sound pressures can be greatly reduced in the entire frequency range investigated. The normalized relative reconstruction error introduced by the parameters of the hydrophone array and the water tank in the experiment is analyzed. The results of the experimental validation show that the proposed approach is promising for accurately measuring sound signals radiated from manned or unmanned vehicles or equipment underwater near a water-air interface. [ABSTRACT FROM AUTHOR]

The underwater broadband anechoic coating is highly demanded in acoustic stealth for underwater vehicles. In this work, we propose a gradient index metamaterial for broadband underwater sound absorption, which is further experimentally validated under high pressure conditions. The gradient index metamaterial consisting of radius-varied cavities in rubber allows the sound waves to follow the objective trajectory in a broad band with impedance matching at the interface with water. The viscosity of the materials and the coupling effect among the gradient cavities contribute to broadband sound absorption. The performance of broadband sound absorption from 1 to 10 kHz is validated experimentally, which agrees well with the theoretical and numerical results. We further experimentally demonstrate the sound absorption performance with hydrostatic pressure up to 3 MPa and analyze the mechanism of the sound absorption deviation caused by high pressures. The proposed gradient index metamaterial provides a simple and efficient way to develop underwater broadband acoustic coatings. [ABSTRACT FROM AUTHOR]

Organisms, such as fishes and invertebrates and including their larval stages, listen to underwater soundscapes to detect information about nearby habitats. Such soundscapes may be influenced by habitat degradation or enhancement, which can lead to acoustically mediated feedback loops affecting the overall ecosystem.Despite the importance of underwater sounds on ecological functioning, there have been limited studies documenting soundscapes of intertidal ecosystems and few, if any, of living shoreline soundscapes. Living shorelines would especially benefit from acoustically mediated effects for objectives like encouraging fish and invertebrate settlement.This case study used a Before‐After‐Control‐Impact design to sample soundscapes and nekton (i.e., fishes and mobile macroinvertebrates) at a living shoreline construction and a nearby hardened shoreline in Cedar Key, FL (USA). Diel soundscape patterns and acoustic attenuation at the two sites were also described a year following the living shoreline construction.In the acoustic sampling, the high frequency bands of both shorelines were dominated by invertebrate sounds that were influenced by season, site and time of day, while the low frequency band of the living shoreline was often dominated by a loud anthropogenic sound. About a year after the living shoreline installation—despite similar measured acoustic attenuation at both sites—the living shoreline featured louder sound pressure levels compared to the hardened shoreline, which may be particularly beneficial for promoting foundational species and other organism settlement.These results demonstrate that Gulf of Mexico intertidal habitats may have soundscape differences even within close proximity and that living shorelines may enhance acoustic characteristics in ways beneficial to continued shoreline development. This represents an important step in better understanding the relationships between habitat structures, nekton communities, and their associated soundscapes as well as the application of passive acoustic monitoring to improve coastal management and conservation. [ABSTRACT FROM AUTHOR]

Abstract The superhydrophobicity of submerged surfaces typically pertains to the trapped air film at the liquid–solid interface, subject to wettability transitions from a Cassie–Baxter state to more unstable states that gradually collapse to high retention regimes, which are energetically more favorable. In this work, the dynamic evolution of those transient metastable states is correlated to the underwater acoustic performance of laser textured superhydrophobic surfaces, resolving the dependence of the ultrasound spectral response with the immersion time to capture the genuine contribution of the hierarchical subwavelength morphology, regardless of the air layer effects. Acoustic wave attenuation of the incident ultrasound energy is extensively quantified in transmission, accounting for instantaneous broadband sound blocking (>30 dB) within the spectral range 0.5–1.5 MHz. As a result of the air layer detachment with the immersion time, transmission coefficients increase accordingly, while acoustic fields in reflection unexpectedly evolve toward stealthiness and naïve acoustic camouflage, mostly ascribable to dissipative mechanisms at air layer interfaces. The intrinsic decay of the air layer effect is tentatively determined at different frequencies, since quantitative understanding of the transient lifetime governing underwater surface wettability is critical to design stable superhydrophobic character of laser induced subwavelength metastructures on the most promising acoustic materials – from eco‐friendly natural to artificial.

This paper describes the Santos Basin Underwater Soundscape Monitoring Project (PMPAS-BS), a Brazilian ocean soundscape monitoring initiative. The main objective of the project is to quantify and assess hydroacoustic noise of anthropogenic origin in a large sedimentary basin extending from 23° S to 28° S on the southeastern Brazilian continental margin of the South Atlantic Ocean. Noise associated with oil and gas (O&G) exploration and production activities is the primary target, but this oceanic region also has busy shipping lanes for commercial, military, and fishing vessels. The two main hubs of Brazil’s export and import of goods by sea are located in this region: Santos and Rio de Janeiro ports. The project has three measurement components: mobile monitoring based on gliders and drifting acoustic profilers, fixed shallow-water monitoring based on acoustic measurements at coastal stations near shipping lanes associated with exploration and production activities in the Santos Basin, and fixed oceanic monitoring based on deep-water mooring lines equipped with passive autonomous acoustic recorders near production units, shipping lanes, and areas with lower intensity of O&G activities (pristine or reference sites). Numerical modeling of anthropogenic underwater acoustic noise has also been included as a fourth project component. The PMPAS-BS covers an area of more than 251,000 km2 and uses several instruments with different methods and sensors for acoustic measurements. Its results provide current sound levels over a very large region of the western South Atlantic, both in areas more and less affected by anthropogenic activities.

We propose a contactless and non-closed strategy to mitigate the sound radiated from an underwater source. The problem is formulated as the interaction of the source with different cross-sectional shapes (including ideal zero-cross-section sources, circular-cross-section rods, and rectangular-cross-section rods) and the acoustic superscatterer treated as a concentric cylindrical structure. The acoustic superscatterer consisting of an internal core and a coating made of the double-negative acoustic metamaterial with specific constitutive parameters relies on the coating to virtually magnify its internal core, thus yielding intense multiple scattering with the source. According to the transformation media theory and the method of images, the complete landscape has been presented theoretically and numerically of the remote mitigation of underwater source radiation via an acoustic superscatterer. We demonstrate that by appropriately placing the acoustic superscatterer nearby the source, it can achieve the omnidirectional radiation mitigation stemming from the coherent extinction mechanism. A possible scheme for the practical realization of the acoustic superscatterer is also discussed. This work may facilitate the design and application of remotely placed meta-devices for the manipulation of underwater acoustic waves. [ABSTRACT FROM AUTHOR]

Non-cooperative underwater acoustic signal detection and recognition technologies occupy an important position in the marine communication environment. Traditional methods require threshold adjustments based on the environment, making the application less robust. This article addresses the technical challenges of automatic detection and recognition of non-cooperative underwater communication signals. It utilizes two pre-trained models, SSD(Single Shot Detector)-MobileNet-V2-FPNlite(Feature Pyramid Network lite) and EfficientDet, which can simultaneously perform signal classification and automatic drawing of signal bounding boxes on the time-frequency graph. SSD-MobileNet-V2-FPNlite uses the Linear Bottleneck layers and the Inverted Residual structure, which improve the capture of multi-scale representations of signals. EfficientDet performs bidirectional feature fusion to enhance the connections and information flow between signal features. After fine-tuning with simulated data, SSD-MobileNet-V2-FPNlite achieves high detection accuracy on both simulated and real lake test data without transfer learning. EfficientDet also performs well on simulated data and, after transfer learning, shows high accuracy in recognizing and drawing bounding boxes on the lake test data. Both methods exhibit strong generalization ability, feature fewer model parameters, and have lower computational complexity.

The presence of natural ambient noise interferes with the system for locating and identifying underwater targets. This paper suggests that a Dual-Path Transformation Network (DPTN) reduces ambient noise in underwater acoustic signals. First, the input acoustic signals’ higher-order non-linear features are extracted using a multi-scale convolutional encoder neural network. Second, sub-vectors with the same length are created according to the time dimension from the higher-order non-linear features. The sub-vectors are stitched together to form a three-dimensional tensor. Third, a neural network transformer based on the feed-forward network is constructed. Further, to capture long-term series features and separate the target signal from the noisy signals, the three-dimensional tensor is used as the input of the transformer-based masking network. Finally, overlap-add and transpose are used to obtain discernible target signals. The experimental results verify the effectiveness of the proposed underwater acoustic signal denoising algorithm and demonstrate that the proposed DPRN method can obtain higher output signal-to-noise ratio (SNR) and the scale-invariant signal-to-noise ratio (SI-SNR) compared with the other classical algorithms.

A numerical demonstration is given of the feasibility of a newly designed tunable asymmetric acoustic transmission device, for which the acoustic transmission coefficients for perpendicularly incident left- and right-running waves are different at a particular single frequency within a forbidden zone. The proposed device consists of a hybrid, sonic-crystal-based resonant structure with horizontal periodicity and a one-dimensional periodic rectangular acoustic grating with vertical periodicity. The former structure exhibits a wide bandgap through which only waves with particular combinations of their frequency and propagation direction can tunnel. The direction dependence of the acoustic transmission is based on the geometrical asymmetry of the placement of the grating, and on the narrowband correspondence between the diffraction angle of the grating and the allowed transmission direction through the sonic-crystal-based structure. Angular and spectral tuning are possible by changing the grating and/or the sonic crystal periodicity. The device has potential applications in underwater acoustics for asymmetric transmission systems and secure communication systems. [ABSTRACT FROM AUTHOR]

The low-frequency line spectrum of the radiated noise signals of hydroacoustic targets contains features describing the intrinsic properties of the target that make the target susceptible to exposure. In order to extract the line spectral features of underwater acoustic targets, a method combining image processing and a deep autoencoder network (DAE) is proposed in this paper to enhance the low-frequency weak line spectrum of underwater targets in an extremely low signal-to-noise ratio environment based on the measured data of large underwater vehicles. A Gauss–Bernoulli restricted Boltzmann machine (G–BRBM) for real-value signal processing was designed and programmed by introducing a greedy algorithm. On this basis, the encoding and decoding mechanism of the DAE network was used to eliminate interference from environmental noise. The weak line spectrum features were effectively enhanced and extracted under an extremely low signal-to-noise ratio of 10–300 Hz, after which the reconstruction results of the line spectrum features were obtained. Data from large underwater vehicles detected by far-field sonar arrays were processed and the results show that the method proposed in this paper was able to adaptively enhance the line spectrum in a data-driven manner. The DAE method was able to achieve more than double the extractable line spectral density in the frequency band of 10–300 Hz. Compared with the traditional feature enhancement extraction method, the DAE method has certain advantages for the extraction of weak line spectra. [ABSTRACT FROM AUTHOR]

Stable communication technologies in complex waters are a prerequisite for underwater operations. Underwater acoustic communication is susceptible to multipath interference, while underwater optical communication is susceptible to environmental impact. The underwater electric field communication established based on the weak electric fish perception mechanism is not susceptible to environmental interference, and the communication is stable. It is a new type of underwater communication technology. To address issues like short communication distances and high bit error rates in existing underwater electric field communication systems, this study focuses on underwater electric field communication systems based on direct sequence spread spectrum (DSSS) and binary phase shift keying (BPSK) modulation techniques. To verify the feasibility of the established spread spectrum electric field communication system, static communication experiments were carried out in a swimming pool using the DSSS-based system. The experimental results show that in fresh water with a conductivity of 739 μS/cm, the system can achieve underwater current electric field communication within a 11.2 m range with 10−6 bit errors. This paper validates the feasibility of DSSS BPSK in short-range underwater communication, and compact communication devices are expected to be deployed on underwater robots for underwater operations. [ABSTRACT FROM AUTHOR]

To facilitate wireless communication, ocean exploration, and other military and scientific uses, underwater acoustic sensor networks have been developed. This network's location in a complex and demanding underwater environment increases failure rates and lowers the network's efficiency in terms of hardware and software. As a result, defining the quality of service in wireless sensor networks operating underwater is a developing study field. Supporting QoS is a difficult problem because of resource limits in underwater sensor networks, such as computing power, memory, bandwidth, and power source. The hardware architecture, elements, and structure of underwater sound sensor networks have been described in this study, along with a detailed explanation of the network levels and the difficulties that each layer faces. A summary of different QoS-aware routing methods for underwater sensor networks is also provided, along with a discussion of the QoS requirements for each layer. Finally, based on the crucial quality standards for the routing services in the underwater sensor network, each of the aforementioned protocols has been assessed and contrasted. In the conclusion, the service quality in underwater sensor network routing has been enhanced by the presentation of a routing solution based on multi-cast routing, hybrid clustering, and fuzzy logic. [ABSTRACT FROM AUTHOR]

Accurately identifying sources of flow loss and hydroacoustics and clarifying the mechanism of their generation are crucial for directing the optimal design of efficient and quiet pump-jet propulsors (PJPs). In this paper, numerical simulations of steady and unsteady flow are performed for a PJP equipped with pre-swirl stationary vanes, based on which both sources of flow loss and hydroacoustics are investigated at multi-level granularity. Analyses of flow efficiency and entropy generation rate are performed to identify the sources of flow loss, and analyses of thrust fluctuation and wall pressure fluctuation are conduced to identify the sources of hydroacoustics. The results indicate that the pressure drag accounts for 76% of the total drag and is mainly contributed from the stator and the duct, but the flow efficiency of the rotor is much smaller than that of the stator and the sources of the flow loss are mainly located at three regions of the rotating blades: the leading edge, the tip, and the corner of the suction surface. The hydroacoustic sources are mainly located at the leading edge and the tip of the rotating blades due to stator–rotor and duct–rotor interactions, respectively, but the Taylor's frozen turbulence hypothesis is inappropriate to describe the wake evolution of the stationary vanes owing to the potential interaction caused by the blade rotation. [ABSTRACT FROM AUTHOR]

Bubble entrainment caused by and underwater sound produced by a single water drop falling on the surface of freshwater and saltwater were experimentally investigated in this study. A charge-coupled device camera was employed to photograph the drop's impact and the process of bubble formation. Underwater sound was received using a hydrophone connected to a charge amplifier and a data-acquisition system. To capture the temporal and spectral characteristics of the sound, the time series of the underwater sound was transformed into the time–frequency domain by using the Gabor transform. The diameter of the drops was varied from 2.6 to 5.66 mm, and the release heights investigated were 10–150 cm. The experimental results revealed three distinct bubble formation processes following the impact of a water drop. Type-I bubbling was the formation of a singular bubble at the apex of the contracting crater produced by the initial drop impact. Type-II bubbling was the formation of one or multiple bubbles entrained by the downward motion of the water column produced during the initial drop impact. Type-III bubbling was the formation of a singular bubble generated due to the impact of a secondary water drop above the water column. The effects of salinity on bubble entrainment and underwater noise were also studied. For the same experimental parameters, bubbles entrained in saltwater were discovered to be larger than those entrapped in freshwater; moreover, the saltwater bubbles collapsed into two bubbles. [ABSTRACT FROM AUTHOR]

Underwater Acoustic Target Recognition (UATR) plays a crucial role in underwater detection devices. However, due to the difficulty and high cost of collecting data in the underwater environment, UATR still faces the problem of small datasets. Few-shot learning (FSL) addresses this challenge through techniques such as Siamese networks and prototypical networks. However, it also suffers from the issue of overfitting, which leads to catastrophic forgetting and performance degradation. Current underwater FSL methods primarily focus on mining similar information within sample pairs, ignoring the unique features of ship radiation noise. This study proposes a novel cross-domain contrastive learning-based few-shot (CDCF) method for UATR to alleviate overfitting issues. This approach leverages self-supervised training on both source and target domains to facilitate rapid adaptation to the target domain. Additionally, a base contrastive module is introduced. Positive and negative sample pairs are generated through data augmentation, and the similarity in the corresponding frequency bands of feature embedding is utilized to learn fine-grained features of ship radiation noise, thereby expanding the scope of knowledge in the source domain. We evaluate the performance of CDCF in diverse scenarios on ShipsEar and DeepShip datasets. The experimental results indicate that in cross-domain environments, the model achieves accuracy rates of 56.71%, 73.02%, and 76.93% for 1-shot, 3-shot, and 5-shot scenarios, respectively, outperforming other FSL methods. Moreover, the model demonstrates outstanding performance in noisy environments. [ABSTRACT FROM AUTHOR]

This study proposes a novel application of neural computing based on deep learning for the real-time prediction of motion parameters for underwater maneuvering object. The intelligent strategy utilizes the capabilities of Scaled Conjugate Gradient Neural Intelligence (SCGNI) to estimate the dynamics of underwater target that adhere to discrete-time Markov chain. Following a state-space methodology in which target dynamics are combined with noisy passive bearings, nonlinear probabilistic computational algorithms are frequently used for motion parameters prediction applications in underwater acoustics. The precision and robustness of SCGNI are examined here for effective motion parameter prediction of a highly dynamic Markov chain underwater passive vehicle. For investigating the effectiveness of the soft computing strategy, a steady supervised maneuvering route of undersea passive object is designed. In the framework of bearings-only tracking technology, system modeling for parameters prediction is built, and the effectiveness of the SCGNI is examined in ideal and cluttered marine atmospheres simultaneously. The real-time location, velocity, and turn rate of dynamic target are analyzed for five distinct scenarios by varying the standard deviation of white Gaussian observed noise in the context of mean square error (MSE) between real and estimated values. For the given motion parameters prediction problem, sufficient Monte Carlo simulation results support SCGNI's superiority over typical generalized pseudo-Bayesian filtering strategies such as Interacting Multiple Model Extended Kalman Filter (IMMEKF) and Interacting Multiple Model Unscented Kalman Filter (IMMUKF). [ABSTRACT FROM AUTHOR]

The rarest seal and the world's most endangered pinniped species, the Mediterranean monk seal (Monachus monachus), has a small and isolated population in the Madeira Archipelago (Portugal). This species tends to be extremely wary of humans and, therefore, very difficult to approach and study.Passive acoustic monitoring (PAM) is a non‐invasive, cost‐effective tool that can be a valuable complement for the traditional monitoring methods, providing insight for effective conservation of the seal in the Madeira Archipelago.In this pilot study, custom‐designed autonomous underwater recorders were deployed in two marine protected areas (Garajau Partial Nature Reserve and the Desertas Islands Nature Reserve) to assess the potential of PAM to detect and monitor this elusive and endangered species in the Madeira Archipelago.Two call types putatively produced by M. monachus were detected in a subsample of audio files recorded over a 3‐month acoustic deployment; these call types share similarities with the /growl/ and /hiccup/ recently described for M. monachus in a Mediterranean population. The most common sound type detected was the low‐frequency growl. No obvious pattern was found in the abundance of sounds according to sampling date, and no significant difference was found in the abundance of sounds in different periods of the day.The ability to detect the species' underwater vocalizations with PAM opens the possibility of future monitoring plans based on data obtained from audio recordings. These data can provide relevant information for conservation, namely, on the presence and abundance of the seals. [ABSTRACT FROM AUTHOR]

Improving underwater image quality is crucial for marine detection applications. However, in the marine environment, captured images are often affected by various degradation factors due to the complexity of underwater conditions. In addition to common color distortions, marine snow noise in underwater images is also a significant issue. The backscatter of artificial light on marine snow generates specks in images, thereby affecting image quality, scene perception, and subsequently impacting downstream tasks such as target detection and segmentation. Addressing the issues caused by marine snow noise, we have designed a new network structure. In this work, a novel skip-connection structure called a dual channel multi-scale feature transmitter (DCMFT) is implemented to reduce information loss during downsampling in the feature encoding and decoding section. Additionally, in the feature transfer process for each stage, iterative attentional feature fusion (iAFF) modules are inserted to fully utilize marine snow features extracted at different stages. Finally, to further optimize the network's performance, we incorporate the multi-scale structural similarity index (MS-SSIM) into the loss function to ensure more effective convergence during training. Through experiments conducted on the Marine Snow Removal Benchmark (MSRB) dataset with an augmented sample size, our method has achieved significant results. The experimental results demonstrate that our approach excels in removing marine snow noise, with a peak signal-to-noise ratio reaching 38.9251 dB, significantly outperforming existing methods. [ABSTRACT FROM AUTHOR]

Vibration induced by marine power devices (MPD) transmitting to the hull structure is one of the most important factors that cause ship vibration and underwater sound radiation. Vibration isolation technologies (VIT) are widely applied to reduce the vibration transmission. However, the overweight issue of VIT for marine power devices is a currently challenging engineering problem. The current reserve of lightweight and high-efficiency VIT for MPD and relevant theoretical and design research are seriously insufficient. This article first elaborates the causes of the overweight problem of VIT for MPD: (1) failing to grasp the quantitative law; (2) single vibration suppression mechanism. Then, it systematically sorts out the technical methods and application examples with potential to solve the overweight problem, such as dynamic optimization design, lightweight material method, novel intermediate mass structures, distributed dynamic vibration absorbers (DDVAs), locally resonant structures (LRS), particle damping (PD), quasizero stiffness isolators (QZSI), and active vibration control (AVC) technologies. Finally, the future development of lightweight VIT for MPD is prospected. It can be used as a reference for marine vessel vibration attenuation research and engineering design. [ABSTRACT FROM AUTHOR]

Underwater Sound Speed Profile (SSP) distribution is crucial for the propagation mode of acoustic signals, so fast and accurate estimation of SSP is of great importance in building underwater observation systems. The state-of-the-art SSP inversion methods include frameworks of matched field processing (MFP), compressive sensing (CS), and feed-forward neural networks (FNNs), among which the FNN shows better real-time performance while maintaining the same level of accuracy. However, the training of FNN needs quite a lot historical SSP samples, which is difficult to satisfy in many ocean areas. This situation is called few-shot learning. To tackle this issue, we propose a multi-task learning (MTL) model with partial parameter sharing among different training tasks. By MTL, common features could be extracted, which accelerates the learning process on given tasks, and reduces the demand for reference samples, enhancing the generalization ability in few-shot learning. To verify the feasibility and effectiveness of MTL, a deep-ocean experiment was held in April 2023 in the South China Sea. Results show that MTL outperforms the other mainstream methods in terms of accuracy for SSP inversion, while inheriting the real-time advantage of FNN during the inversion stage. [ABSTRACT FROM AUTHOR]

This article presents the investigation of the wind-driven component of underwater ambient sound in the shallow brackish waters of the Baltic Sea. Natural sound levels are strongly correlated with the wind speed at high frequencies (≥5 kHz). At frequencies above 5 kHz, a characteristic spectral level decrease of 5 dB/octave was observed. Analysis of the data revealed that, for the same wind speed, the spectral levels are higher when the wind is blowing from a direction where the closest obstructing shore is farther away, i.e. the wind fetch is longer, which results in higher waves. This was especially noticeable in ambient sound recorded in a channel-like basin, where for the 8.5 m/s steady wind speed, the 5 kHz mean spectral level is 4 dB higher at the longer 152 km wind fetch versus the shorter 2.1 km wind fetch. [ABSTRACT FROM AUTHOR]