Benthic habitat mapping is a rapidly growing field of underwater remote sensing studies. This study provides the first insight for high-resolution hydroacoustic surveys in the Slupsk Bank Natura 2000 site, one of the most valuable sites in the Polish Exclusive Zone of the Southern Baltic. This study developed a quick and transparent, automatic classification workflow based on multibeam echosounder and side-scan sonar surveys to classify benthic habitats in eight study sites within the Slupsk Bank. Different predictor variables, four supervised classifiers, and the generalisation approach, improving the accuracy of the developed model were evaluated. The results suggested a very high significance for the classification performance of specific geomorphometric features that were not used in benthic habitat mapping before. These include, e.g., Fuzzy Landform Element Classification, Multiresolution Index of the Valley Bottom Flatness, and Multiresolution Index of the Ridge Top Flatness. Comparison of classification results with manual maps demonstrated that Random Forest had the highest performance of four tested supervised classifiers. Because the current needs include benthic habitat mapping for the whole area of the Polish Exclusive Economic Zone, the key findings of this study may be further applied to extensive areas in the Polish waters and other vast areas worldwide.

The digital representation of seafloor, a challenge in UNESCO’s Ocean Decade initiative, is essential for sustainable development support and marine environment protection, aligning with the United Nations’ 2030 program goals. Accuracy in seafloor representation can be achieved through remote sensing measurements, including acoustic and laser sources. Ground truth information integration facilitates comprehensive seafloor assessment. The current seafloor mapping paradigm benefits from the object-based image analysis (OBIA) approach, managing high-resolution remote sensing measurements effectively. A critical OBIA step is the segmentation process, with various algorithms available. Recent artificial intelligence advancements have led to AI-powered segmentation algorithms development, like the Segment Anything Model (SAM) by META AI. This paper presents the SAM approach’s first evaluation for seafloor mapping. The benchmark remote sensing dataset refers to Puck Lagoon, Poland and includes measurements from various sources, primarily multibeam echosounders, bathymetric lidar, airborne photogrammetry, and satellite imagery. The SAM algorithm’s performance was evaluated on an affordable workstation equipped with an NVIDIA GPU, enabling CUDA architecture utilization. The growing popularity and demand for AI-based services predict their widespread application in future underwater remote sensing studies, regardless of the measurement technology used (acoustic, laser, or imagery). Applying SAM in Puck Lagoon seafloor mapping may benefit other seafloor mapping studies intending to employ AI technology.

The rise of Artificial Neural Networks (ANNs) has revolutionized various fields of research, demonstrating their effectiveness in solving complex problems. However, there are still unexplored areas where the application of neural networks, particularly Convolutional Neural Network (CNN) models, has yet to be explored. One area where the application of Artificial Neural Networks is even expected is geomorphology. One of the tasks of geomorphology is the classification of landforms in a broad sense. Such classification requires a precise interpretation approach to create a homogeneous product, and this requires time and a uniform, consistent approach by the interpreter, which is not easy to achieve with manual operations. Classifications in geomorphology are mainly performed by manual or semi-automatic methods. The use of Artificial Neural Networks can complement and, in many areas, replace manual classification and reduce the time commitment of the interpreter, not least because of its repeatability and objectivity, which is a definite advantage in the case of geomorphological interpretation of vast areas. This study uses two popular CNN architectures, including VGG and Residual Neural Network (ResNet), to solve the problem of classifying glacial and fluvioglacial landforms based on a Digital Elevation Model (DEM). The results of this study show that CNNs can produce high accuracy scores (up to 87% overall accuracy) for a ground-based dataset and are a suitable method for identifying glacial and fluvioglacial landforms using DEM data as input. The presented method provides an objective, reproducible and fast tool for automatic geomorphological analysis of terrain imagery of vast areas.

The aim of this study is to determine the state and changes of the Vistula Canal (Wisła Przekop) external delta relief, in the period of 2009-2014, and in relation to the assessment of the conditions of free runoff of the Vistula River waters. This is an important element for assessing the effectiveness of flood prevention in the area of Żuławy. The basis of the study are detailed bathymetry monitoring and morphometric measurements of the external delta with the supply channel. Modern equipment was used (mainly RTK and DGPS specialist navigation systems and multibeam echosounder), providing the obtainment of a digital model of the seabed surface, with decimeter accuracy of location points. In subsequent years, a series of bathymetric maps for the scale of 1:10 000 was made, as well as relief maps and maps of differential changes of the relief. The observed significant changes in the nature of the relief (with relatively small changes in the contours of the delta), indicate their important role in shaping the conditions of free runoff Vistula waters.Instytut Morski w GdańskuPatryk Sitkiewic

The Southern Baltic seabed is mainly composed of marine sediments situated on glacial and fluvioglacial deposits. In the Kołobrzeg region, however, solid rocks are located at the shallow subbottom. In order to determine the exact distribution of the rocks, some analyses of sonar images, underwater photos, sediment samples, seismoacoustic profiles and bathymetric data have been performed. The outcrops of Jurassic sandstones and mudstones have been documented. The geological structure of the area is presented, while the fact that the present seafloor relief is determined by the shallow uneven top surface of the Jurassic rocks is emphasized. The seabed has been classified as rock platform-constrained. The seafloor is a sediment-poor abrasion platform, within which marine sandy sediments are transported, mainly toward offshore, on the firm, stiff or hard surface of glacial deposits and Jurassic solid rocks. This results in a lack of material for the reconstruction of the beach after storms. Many of the commonly used methods for studying the condition and changes of the coast cannot be used in this type of terrain due to the presence of a thin and discontinuous dynamic layer lying on the erosion-resistant rocks of the geomorphologically complex seafloor.