Growing concern over anthropogenic underwater noise, highlighted by initiatives like the Marine Strategy Framework Directive (MSFD) and its Technical Group on Underwater Noise (TG Noise), emphasizes regions like the Western Black Sea, where increasing activities threaten marine habitats. This region is experiencing rapid growth in maritime traffic and resource exploitation, which is intensifying concerns over the noise impacts on its unique marine habitats. While machine learning offers promising solutions, a research gap persists in comprehensively evaluating diverse ML models within an integrated framework for complex underwater acoustic data, particularly concerning real-world data limitations like class imbalance. This paper addresses this by presenting a multi-faceted framework using passive acoustic monitoring (PAM) data from fixed locations (50–100 m depth). Acoustic data are processed using advanced signal processing (broadband Sound Pressure Level (SPL), Power Spectral Density (PSD)) for feature extraction (Mel-spectrograms for deep learning; PSD statistical moments for classical/unsupervised ML). The framework evaluates Convolutional Neural Networks (CNNs), Random Forest, and Support Vector Machines (SVMs) for noise event classification, alongside Gaussian Mixture Models (GMMs) for anomaly detection. Our results demonstrate that the CNN achieved the highest classification accuracy of 0.9359, significantly outperforming Random Forest (0.8494) and SVM (0.8397) on the test dataset. These findings emphasize the capability of deep learning in automatically extracting discriminative features, highlighting its potential for enhanced automated underwater acoustic monitoring.

The work involves assessing of the marine current regime in the north west of the Black Sea, Midia Cape-Constanta area. The area of interest is chosen due its existing features, using marine currents data (direction and speed) within the CFD to determine the distribution of velocities and pressures. The study on the distribution of velocities and pressures, according to local hydro-meteorological characteristic elements, can be used for identifying the risk areas for navigation and existing hydro-technical construction industry. Modelling results using the CFD analysis shows the currents velocity values, pressures and turbulences of the investigated marine region. The model can be used in the analysis and the coastline changes as well as in risk assessment for hydraulic structures. Additional parameters into the model, such as the sea wave characteristics, aim to provide an indication of the potential hydro-dynamic power that is available at a determined location of interest.

This study explores the use of Temporal Fusion Transformers (TFTs), an AI/ML technique, to enhance the prediction of coastal dynamics along the Western Black Sea coast. We integrate in-situ observations from five meteo-oceanographic stations with modelled geospatial marine data from the Copernicus Marine Service. TFTs are employed to refine predictions of shallow water dynamics by considering atmospheric influences, with a particular focus on wave-wind correlations in coastal regions. Atmospheric pressure and temperature are treated as latitude-dependent constants, with specific investigations into extreme events like freezing and solar radiation-induced turbulence. Explainable AI (XAI) is exploited to ensure transparent model interpretations and identify key influential input variables. Data attribution strategies address missing data concerns, while ensemble modelling enhances overall prediction robustness. The models demonstrate a significant improvement in prediction accuracy compared to traditional methods. This research provides a deeper understanding of atmosphere-marine interactions and demonstrates the efficacy of Artificial intelligence (AI)/Machine Learning (ML) in bridging observational and modelled data gaps for informed coastal zone management decisions, essential for maritime safety and coastal management along the Western Black Sea coast.

Coastal erosion poses a significant risk to the Romanian Black Sea coast, a region characterized by the interaction of natural geomorphological processes and anthropogenic pressures. The research focuses on developing a tool to quantify the cumulative impact of hydro-geo-morphological factors and to assess the vulnerability of the coastal zone to these influences. The approach involves adapting the Coastal Vulnerability Index (CVI)—previously applied in various methodologies—to the specific characteristics of this semi-enclosed basin, which included the exclusion of the tidal range variable due to the Black Sea’s negligible tidal amplitude. The selection of key variables, including coastal geology and geomorphology, shoreline change rates, coastal slope, sea level, and wave regime, was conducted with consideration for the specific characteristics of the Romanian coastal zone. By classifying these variables on a semi-quantitative scale and integrating them into a CVI, the study identifies and maps areas of high vulnerability. The analysis, based on a 1 × 1 km grid resolution, identified sectors of very high vulnerability in the northern Danube Delta unit, particularly along the coastlines of South Sulina–Câşla Vădanei, Sahalin, and Periboina-Edighiol-Vadu. These findings are validated through a comparison with long-term, multidecadal shoreline evolution data, confirming the model’s predictive accuracy. While the 1 × 1 km grid is effective for a macro-scale assessment, the study highlights the need for a finer resolution (e.g., 100 × 100 m) for detailed analysis in the southern region, due to localized geodynamic conditions and the significant influence of coastal infrastructure.

Global Navigation Satellite Systems (GNSSs) are the primary source of information for Positioning, Navigation, and Timing (PNT) in the maritime sector; however, they are vulnerable to unintentional or deliberate interference, such as jamming, spoofing, or meaconing. The continuous monitoring of GNSS signals is crucial for vessels and mobile maritime platforms to ensure the integrity, availability, and accuracy of positioning and navigation services. This monitoring is essential for guaranteeing the safety and security of navigation and contributes to the accurate positioning of vessels and platforms involved in hydrographic and oceanographic research. This paper presents the implementation of a complex system for monitoring the quality of signals within the GNSS spectrum at the Maritime Hydrographic Directorate (MHD). The system provides real-time analysis of signal parameters from various GNSSs, enabling alerts in critical situations and generating statistics and reports. It comprises four permanent stations equipped with state-of-the-art GNSS receivers, which integrate a spectrum analyzer and store raw data for post-processing. The system also includes software for monitoring the GNSS spectrum, detecting interference events, and visualizing signal quality data. Implemented using a Docker-based platform to enable efficient management and distribution, the software architecture consists of a reverse proxy, message broker, front-end, authorization service, GNSS orchestrator, and GNSS monitoring module. This system enhances the quality of command, control, communications, and intelligence decisions for planning and execution. It has demonstrated a high success rate in detecting and localizing jamming and spoofing events, thereby improving maritime situational awareness and navigational safety. Future development could involve installing dedicated stations to locate interference sources.

Microplastic pollution presents a significant and rising risk to both ecological integrity and the long-term viability of economic activities reliant on marine ecosystems. The Black Sea, a region sustaining economic sectors such as fisheries, tourism, and maritime transport, is increasingly vulnerable to this form of contamination. Mytilus galloprovincialis, a well-established bioindicator, accumulates microplastics, providing a direct measure of environmental pollution and indicating potential economic consequences deriving from degraded ecosystem services. While previous studies have documented microplastic pollution in the Black Sea, our paper specifically quantified microplastic contamination in M. galloprovincialis collected from four sites along the western Black Sea coast, each characterised by distinct levels of anthropogenic influence: Midia Port, Constanta Port, Mangalia Port, and 2 Mai. We used statistical analysis to quantify site-specific microplastic contamination in M. galloprovincialis and employed machine learning to develop models predicting accumulation patterns based on environmental variables. Our findings demonstrate the efficacy of mussels as bioindicators of marine plastic pollution and highlight the utility of machine learning in developing effective predictive tools for monitoring and managing marine litter contamination in marine environments, thereby contributing to sustainable economic practices.