In austral winter, biological productivity at the Angolan shelf reaches its maximum. The alongshore winds, however, reach their seasonal minimum suggesting that processes other than local wind‐driven upwelling contribute to near‐coastal cooling and upward nutrient supply, one possibility being mixing induced by internal tides (ITs). Here, we apply a three‐dimensional ocean model to simulate the generation, propagation, and dissipation of ITs at the Angolan continental slope and shelf. Model results are validated against moored acoustic Doppler current profiler and other observations. Simulated ITs are mainly generated in regions with a critical/supercritical slope typically between the 200‐ and 500‐m isobaths. Mixing induced by ITs is found to be strongest close to the coast and gradually decreases offshore thereby contributing to the establishment of cross‐shore temperature gradients. The available seasonal coverage of hydrographic data is used to design simulations to investigate the influence of seasonally varying stratification characterized by low stratification in austral winter and high stratification in austral summer. The results show that IT characteristics, such as their wavelengths, sea surface convergence patterns, and baroclinic structure, have substantial seasonal variations and additionally strong spatial inhomogeneities. However, seasonal variations in the spatially averaged generation, onshore flux, and dissipation of IT energy are weak. By evaluating the change of potential energy, it is shown, nevertheless, that mixing due to ITs is more effective during austral winter. We argue that this is because the weaker background stratification in austral winter than in austral summer acts as a preconditioning for IT mixing.

In a SAR image acquired by the ERS-2 satellite, crossed “X-shape” internal solitary waves (ISWs) south to Dongsha Island are found to be a wave–wave interaction composed of five solitons: two head waves, two tail waves, and the overlapped part. To explain this remote sensing phenomenon, based on a high-resolution three-dimensional MIT general circulation model (MITgcm) using realistic topography and tidal forcing, the “X-shape” internal waves are reproduced at the same location. The development processes of the waves indicate that the “X-shape” ISWs are two waves diffracted from one internal wave southeast to Dongsha Island. During the propagation, the amplitude of their overlapped part of the “X-shape” ISWs becomes significantly larger than the sum of the amplitudes of both head waves, which proves that nonlinear wave–wave interaction has occurred. Based on wave–wave interaction theory, the theoretical maximum value of the amplitude of the overlapped part at the initial moment is calculated as 14.12 m, which is in good agreement with the model results of 14 m. Meanwhile, the variation of the theoretical amplitude of the overlapped part is basically consistent with that of the modeled one, confirming the occurrence of the wave–wave interaction. Besides, when the waves propagate over varying water depths, the type of the wave–wave interaction can change rather than being fixed from the start.

Ocean internal waves occur in stably stratified seawater and play a crucial role in energy cascade, material transport, and military activities. However, the complex and irregular spatial patterns of internal waves pose significant challenges for accurate detection in SAR images when using conventional convolutional neural networks, which often lack adaptability to geometric variations. To address this problem, this paper proposes a refined Faster R-CNN detection framework, termed “rFaster R-CNN”, and adopts a transfer learning strategy to enhance model generalization and robustness. In the feature extraction stage, a backbone network called “ResNet50_CDCN” that integrates the CBAM attention mechanism and DCNv2 deformable convolution is constructed to enhance the feature expression ability of key regions in the images. Experimental results show that in the internal wave dataset constructed in this paper, this network improves the detection accuracy by approximately 3% compared to the original ResNet50 network. At the region proposal stage, this paper further adds two small-scale anchors and combines the ROI Align and FPN modules, effectively enhancing the spatial hierarchical information and semantic expression ability of ocean internal waves. compared with classical object detection algorithms such as SSD, YOLO, and RetinaNet, the proposed “rFaster R-CNN” achieves superior detection performance, showing significant improvements in both accuracy and robustness.

In austral winter, biological productivity at the Angolan shelf reaches its maximum. The alongshore winds, however, reach their seasonal minimum suggesting that processes other than local wind-driven upwelling contribute to near-coastal cooling and nutrient supply, one possibility being mixing induced by internal tides (ITs). Here, we apply a three-dimensional ocean model to simulate the generation, propagation and dissipation of ITs at the Angolan slope and shelf. Model results are validated against moored acoustic Doppler current profiler and other observations. Simulated ITs are mainly generated in regions with a critical/supercritical slope typically between the 200- and 500-m isobaths. Mixing induced by ITs is found to be strongest close to the shore and gradually decreases offshore thereby contributing to the establishment of a cross-shore sea surface temperature gradient. The available seasonal coverage of hydrographic data is used to design sensitivity simulations. Seasonal variations in stratification results in substantial temporal differences in IT characteristics, such as their wavelengths, sea surface convergence patterns and baroclinic structure, additionally showing strong spatial variations. However, seasonal variations in the domain-integrated generation, onshore flux and dissipation of IT energy are weak. By defining a parameter - the relative change of the vertical density gradient - to evaluate the relative mixing strength, it is shown, nevertheless, that mixing due to ITs is more effective at weakening the stratification during austral winter. We argue this is because less energy is required to mix the water column in austral winter than in austral summer.

In austral winter, biological productivity at the Angolan shelf reaches its maximum. The alongshore winds, however, reach their seasonal minimum suggesting that processes other than local wind-driven upwelling contribute to near-coastal cooling and upward nutrient supply, one possibility being mixing induced by internal tides (ITs). Here, we apply a three-dimensional ocean model to simulate the generation, propagation and dissipation of ITs at the Angolan continental slope and shelf. Model results are validated against moored acoustic Doppler current profiler and other observations. Simulated ITs are mainly generated in regions with a critical/supercritical slope typically between the 200- and 500-m isobaths. Mixing induced by ITs is found to be strongest close to the coast and gradually decreases offshore thereby contributing to the establishment of cross-shore temperature gradients. The available seasonal coverage of hydrographic data is used to design simulations to investigate the influence of seasonally varying stratification characterized by low stratification in austral winter and high stratification in austral summer. The results show that IT characteristics, such as their wavelengths, sea surface convergence patterns and baroclinic structure, have substantial seasonal variations and additionally strong spatial inhomogeneities. However, seasonal variations in the spatially-averaged generation, onshore flux and dissipation of IT energy are weak. By evaluating the change of potential energy, it is shown, nevertheless, that mixing due to ITs is more effective during austral winter. We argue this is because the weaker background stratification in austral winter than in austral summer acts as a preconditioning for IT mixing.