Journal of Marine Science and Engineering

This study investigated the physicochemical properties and nutrient dynamics on the Romanian shelf of the northwestern Black Sea in July 2024, collecting data across 36 stations (13–1116 m depth) heavily influenced by Danube discharges. Vertical CTD profiling revealed a pronounced seasonal thermocline and a deep-lying permanent halocline. The Cold Intermediate Layer (CIL) boundary, defined by the 8 °C isotherm, was absent, indicating warmer subsurface waters. Surface nutrient concentrations, particularly for nitrate (NO₃) and phosphate (PO₄), were considerably lower than peak eutrophication periods, approaching pre-1970s values, suggesting a positive trend due to reduced anthropogenic loading. They are also comparable to or lower than other coastal regions in the Black Sea. Vertical nutrient profiles confirmed the typical anoxic Black Sea structure, but with regional specifics: the PO₄ maximum was slightly deeper, and the NO₃ maximum position and concentration mirrored the pre-eutrophication period, further supporting reduced anthropogenic nitrogen input. Silicate (SiO₄) concentrations were consistently low throughout the water column, suggesting the northwest shelf functions as a SiO₄ sink compared to the southeastern Black Sea. Overall results indicate a shift towards a less eutrophic state on the Romanian shelf while highlighting the continued dominance of Danube-driven hydrodynamics. In addition to those investigations, this study assessed nutrient preservation techniques, finding that pasteurization was significantly superior to freezing for maintaining the stability of PO₄ and NOx (losses up to 20% and 47% for frozen samples, respectively) over six months. Though SiO₄ was stable under both methods, the freezing produced lower concentrations, possibly from incomplete depolymerization during thawing. These findings stress that pasteurization could be taken into consideration as a reliable preservation technique for long-term storage of nutrient samples.

Wind turbine tower structures composed of slender steel cylindrical shells mainly serve as primary load-bearing components and can be particularly susceptible to buckling due to their thin walls. Ensuring the structural safety of wind turbines therefore requires a clear understanding of the behavior of slender cylindrical shells, which is influenced by material properties, boundary conditions, and loading scenarios. This study experimentally investigates the structural responses of scaled cylindrical structures representing wind turbine towers beyond the proportional limit including the ultimate and post-ultimate strength depending on boundary conditions (fully and frictionally supported). Lateral loads were applied at the top of the specimens to simulate concentrated loads transferred from wind forces on the blades. Furthermore, a numerical model was developed to analyze the structural behavior of the tower validated against the experimental test results. The results provide valuable insights into optimizing the structural design of both onshore and offshore wind turbine towers, contributing to enhanced safety and performance under varying load conditions.

Marine shipping safety is of great concern to many stakeholders, especially maritime authorities, and the consequences of marine accidents, linked to the accident severity and type, are intrinsically impacted by various risk influence factors (RIFs). To investigate the impacts of RIFs on marine accidents and the consequences thereof within Chinese waters, in this study, 1106 marine accident investigation reports issued by China’s MSA during the 2013–2024 period were collected, and a database of marine shipping RIFs was developed based on these data. As a result, 14 typical features were extracted, and the accident severity level and accident type were set as the output features. Then, a comprehensive machine learning algorithm integrating squeeze-and-excitation (SE), a convolutional neural network (CNN), and a gated recurrent unit (GRU) was proposed to process the extracted marine RIFs. Finally, these features were analyzed in terms of importance, correlation, and partial dependence plots (PDPs), and the performance of the SE-CNN-GRU algorithm, especially the prediction accuracy, was verified. The findings and results obtained from this study are valuable for improving shipping safety in Chinese waters; managerial implications are additionally proposed.