18 January 2021 in IEEE Access
Multi-beam echo sounders (MBESs) are characterized by the high resolution and high density of the sounding data. The processing of MBES bathymetry data is of special interest currently in marine surveying. The Combined Uncertainty and Bathymetry Estimator (CUBE) and surface filtering are the main MBES-processing algorithms for outliers. These algorithms involve five adjustable parameters; however, few studies have looked at parameter optimization. In this paper, a Parameter Group Optimization (PGO) method that determines the optimal parameters of CUBE and surface filtering based on the seafloor topographic characteristics of the survey area is presented. The method includes typical area selection, optimal grid resolution analysis, parameter group testing and batch processing, sounding and grid analysis. Raw MBES datasets from shallow- and deep-water survey areas (between 10 and 11000 m deep) are used to validate the proposed method. The results show that when the optimized parameters are used in the CUBE and filtering algorithm, the outliers are automatically eliminated; the processed bathymetry data is in good agreement with the bathymetry derived by a traditional manual processing method, while the processing efficiency can be improved by more than 8 times.
30 September 2019 in IEEE Access
Access to reliable and accurate bathymetric data is fundamental to many marine activities. This paper proposes a merge-normalization (MN) method that is suitable for multi-source bathymetric data fusion in deep ocean areas, to solve the problem of difficult to integrate high-precision digital bathymetric model (DBM) for complex sources and various resolutions of global deep ocean bathymetric data. Then we apply it to the DBM construction of the Mariana Trench. The method combines multi-beam, single-beam, and electronic navigational chart data with Shuttle Radar Topography Mission (SRTM) dataset by using the workflow of merging and normalizing, which can fill the data gaps while preserving topographic details in high-resolution bathymetric data. Compared with the widely used General Bathymetric Chart of the Oceans (GEBCO) dataset, the Mariana Trench dataset constructed in this study demonstrated improved accuracy, resolution, and topographic detail, highlighting the value of the application of the method and of its development potential.
26 May 2023 in IEEE Transactions on Geoscience and Remote Sensing
The coastal zone is connected to the hinterland of the basin and the wide sea, which is not only affected by complex natural factors, but also by human activities. The study of sediment classification in this area will help to further explain topographic evolution and dynamic mechanism. Aiming at the complex mixed sediment environment in the ancient valley area of shallow seas, an advanced deep convolutional neural network classification model based on transfer learning—selective kernel hybrid dilated ResNet-50 (SKHD-ResNet-50) is thus proposed and applied to the coastal area in the Firth of Forth, Scotland, UK. The model overcomes problems of low accuracy of sediment classification with small samples, while effectively improving the efficiency of mixed sediment classification. Five transferred convolutional neural network models including CNN, AlexNet, VGG-19, GoogleNet, and SKHD-ResNet-50, were used to conduct seabed sediment classification experiments on multibeam backscatter mosaics in the study area. An ablation study about the effect of selective kernel convolution (SKC) and hybrid dilated convolution (HDC) was also conducted to further prove the usability of the proposed method. Finally, the optimized convolutional neural network was used to build a multi-layer deep learning model for sediment classification, combining the transfer learning method with the neural network to obtain the optimal network weights through deep global optimization, thus accurately mining statistical characteristics and distribution rules of backscatter intensity data. The experimental results show that the advanced model obtains better classification accuracy than that of state-of-art methods, with overall accuracy and Kappa coefficient of 96.87% and 0.9482, respectively. Findings highlight the proposed sediment classification model, which can effectively distinguish mixed sediments, and greatly improve the accuracy and efficiency of sediment classification.
27 March 2026 in Geophysical Research Letters
Regional geodynamic modeling has shown that oceanic plateau subduction can modify the slab behavior. However, few studies have explored these interactions beyond 100 Myr in 3‐D spherical models. Using 3‐D spherical mantle convection models with self‐consistent plate‐like behavior, we investigate how the geometry and rheological properties of oceanic plateaus influence trench evolution and mantle flow across regional to global scales. Our models show that plateau width and thickness exert the strongest control on trench retreat, while buoyancy alone plays a secondary role. A mechanically strong plateau produces a transient arcuate trench morphology that gradually heals once the plateau is fully subducted, leaving only a delay in retreat. The pronounced curvature of the northern Mariana Trench suggests that the Ogasawara Plateau is mechanically strong. Although plateau subduction leaves no long‐term geological scar unless slab break‐off occurs, it substantially slows trench migration and can influence global tectonic evolution for over 200 Myr.