This paper proposes a two-dimensional underwater sound propagation model using the Discontinuous Galerkin Finite Element Method (DG-FEM) to investigate the influence of current on sound propagation. The acoustic field is calculated by the convected wave equation with the current speed parameter. Based on the current speed data from an assimilation model, a two-dimensional coupled acoustic propagation model in the Fram Strait water area is established to observe the variability in propagation loss under different seasonal velocities in the real ocean environment. The transmission loss and signal time structure are examined. The results obtained in different source frequencies are also compared. It appears that the current velocity, time and range variation all have an effect on underwater sound propagation.

Warm current has a strong impact on the melting of sea ice, so clarifying the current features plays a very important role in the Arctic sea ice coverage forecasting study field. Currently, Arctic acoustic tomography is the only feasible method for the large-range current measurement under the Arctic sea ice. Furthermore, affected by the high latitudes Coriolis force, small-scale variability greatly affects the accuracy of Arctic acoustic tomography. However, small-scale variability could not be measured by empirical parameters and resolved by Regularized Least Squares (RLS) in the inverse problem of Arctic acoustic tomography. In this paper, the convolutional neural network (CNN) is proposed to enhance the prediction accuracy in the Arctic, and especially, Gaussian noise is added to reflect the disturbance of the Arctic environment. First, we use the finite element method to build the background ocean model. Then, the deep learning CNN method constructs the non-linear mapping relationship between the acoustic data and the corresponding flow velocity. Finally, the simulation result shows that the deep learning convolutional neural network method being applied to Arctic acoustic tomography could achieve 45.87% accurate improvement than the common RLS method in the current inversion.

The bowhead whale is a vital component of the maritime environment. Using deep learning techniques to recognize bowhead whales accurately and efficiently is crucial for their protection. Marine acoustic remote sensing technology is currently an important method to recognize bowhead whales. Adaptive SWT is used to extract the acoustic features of bowhead whales. The CNN-LSTM deep learning model was constructed to recognize bowhead whale voices. Compared to STFT, the adaptive SWT used in this study raises the SCR for the stationary and nonstationary bowhead whale whistles by 88.20% and 92.05%, respectively. Ten-fold cross-validation yields an average recognition accuracy of 92.85%. The method efficiency of this work was further confirmed by the consistency found in the Beaufort Sea recognition results and the fisheries ecological study. The research results in this paper help promote the application of marine acoustic remote sensing technology and the conservation of bowhead whales.

It is important to distinguish the dominant mechanism of seabed acoustic scattering for the quantitative inversion of seabed parameters. An identification scheme is proposed based on Bayesian inversion with the relative entropy used to estimate dominant acoustic backscatter mechanism. DiffeRential Evolution Adaptive Metropolis is used to obtain samples from posterior probability density in Bayesian inversion. Three mechanisms for seabed scattering are considered: scattering from a rough water-seabed interface, scattering from volume heterogeneities, and mixed scattering from both interface roughness and volume heterogeneities. Roughness scattering and volume scattering are modelled based on Fluid Theories using Small-Slope Approximation and Small-Perturbation Fluid Approximation, respectively. The identification scheme is applied to three simulated observation data sets. The results indicate that the scheme is promising and appears capable of distinguishing sediment volume from interface roughness scattering and can correctly identify the dominant acoustic backscatter mechanism.

Acoustic tracking of whales’ underwater cruises is essential for protecting marine ecosystems. For cetacean conservationists, fewer hydrophones will provide more convenience in capturing high-mobility whale positions. Currently, it has been possible to use two hydrophones individually to accomplish direction finding or ranging. However, traditional methods only aim at estimating one of the spatial parameters and are susceptible to the detrimental effects of reverberation superimposition. To achieve complete whale tracking under reverberant interference, in this study, an intelligent acoustic tracking model (CIAT) is proposed, which allows both horizontal direction discrimination and distance/depth perception by mining unpredictable features of position information directly from the received signals of two hydrophones. Specifically, the horizontal direction is discriminated by an enhanced cross-spectral analysis to make full use of the exact frequency of received signals and eliminate the interference of non-source signals, and the distance/depth direction combines convolutional neural network (CNN) with transfer learning to address the adverse effects caused by unavoidable acoustic reflections and reverberation superposition. Experiments with real recordings show that 0.13 km/MAE is achieved within 8 km. Our work not only provides satisfactory prediction performance, but also effectively avoids the reverberation effect of long-distance signal propagation, opening up a new avenue for underwater target tracking.

The varies of ice floe thickness (IFH) is a very influential research content in the Earth system academic research, it is closely related to the distribution of underwater acoustic fields. However, the discontinuous physical parameters of ice floes make it difficult to clear this relationship with conventional acoustic propagation models or measured data. Unlike the conventional acoustic models that treat ice floes as one or more physical simplified acoustic reflection layers, this study employs machine learning and FEM (finite element method) modeling to incorporate ice floes' physical parameters into the acoustic field model. The physical parameters of ice floes associated with acoustic field modeling are screened by using the principal component analysis method. On this basis, these ice floes were spatially divided into three highly differentiated and interpretable typical ice clusters using the K‐Means method. We treated ice as a multi‐layer elastic medium, and then the FEM acoustic model was built under the fluid‐structure coupling condition. Measured data (UNDER‐ICE) and MIT acoustic models (OASES) were used to validate our FEM results. We demonstrated that the modeling improves the distribution characteristics of surface channel arrival amplitude under ice floes cover. From the perspective of the model application, the acoustic field response as IFH varies is closely related to the category and region of ice floes.