Accurate water-level prediction is a critical component for ensuring safe maritime navigation, optimizing port operations, and mitigating coastal flooding risks. However, the complex, non-linear spatiotemporal dynamics of water systems pose significant challenges for current forecasting models. The proposed framework introduces three key innovations. First, a dual-weight graph construction mechanism integrates geographical proximity with Dynamic Time Warping (DTW)-derived temporal similarity to better represent hydrodynamic connectivity in coastal and estuarine environments. Second, a state-aware weighted loss function is designed to enhance predictive accuracy during critical hydrological events, such as storm surges and extreme tides, by prioritizing the reduction in errors in these high-risk periods. Third, the WS-STGTN architecture combines graph attention with temporal self-attention to capture long-range dependencies in both space and time. Extensive experiments are conducted using water-level data from five stations in the tidal-influenced lower Yangtze River, a vital artery for shipping and a region susceptible to coastal hydrological extremes. The results demonstrate that the model consistently surpasses a range of baseline methods. Notably, the WS-STGTN achieves an average reduction in Mean Squared Error (MSE) of 27.6% compared to the standard Transformer model, along with the highest coefficient of determination ($R^2\approx 0.96$) across all datasets, indicating its stronger explanatory power for observed water-level variability. This work provides a powerful tool that can be directly applied to improve coastal risk management, marine navigation safety, and the operational planning of port and coastal engineering projects.

Accurate vessel trajectory prediction from Automatic Identification System (AIS) records is important for maritime traffic monitoring, route planning, and intelligent vessel traffic management. However, reliable prediction remains challenging for long forecasting horizons and turning maneuvers. To address this problem, this study proposes the Geometry-Consistent Trajectory Transformer (GCT-Former), a progressive and refinement-based framework for AIS-based vessel trajectory prediction. The proposed model integrates multi-scale historical trajectory encoding, progressive residual future-position generation, and global–local trajectory refinement to improve the stability and continuity of long-horizon trajectory prediction. The predicted trajectories are evaluated as geometric future-position estimates and can provide trajectory-level information for downstream maritime traffic monitoring and decision-support applications. Experiments are conducted on three real-world Danish maritime regions: Aarhus Bay, Great Belt, and Skagen. Compared with representative conventional and deep-learning trajectory prediction models, the proposed model shows its most consistent advantage in long-horizon prediction, particularly in terms of ADE and FDE. In the long-term setting, it achieves average displacement errors of 0.344, 0.546, and 0.218 km and final displacement errors of 0.774, 1.368, and 0.525 km on Aarhus Bay, Great Belt, and Skagen, respectively. The ablation analysis further shows that removing the multi-scale encoding module increases the long-term average displacement error by 7%, 4%, and 3%, while removing the progressive residual decoder leads to larger increases of 15%, 9%, and 8% on the three datasets. The turning-maneuver analysis also shows lower geometric prediction errors under mild-turning and sharp-turning scenarios. These results indicate that GCT-Former improves AIS-based vessel trajectory prediction, especially for long-horizon and maneuvering cases.

Freeze-up and breakup dates in cold-region rivers strongly influence winter navigation closure, spring navigation resumption, and navigation safety during ice runs. However, river-ice transitions are reach-specific and cannot be reliably determined from instantaneous air-temperature thresholds alone, while recursive multi-step forecasting is vulnerable to error accumulation. To address these challenges, this study develops the SAG-AL-ESN-HRCLRC framework for multi-station air-temperature forecasting and reach-specific freeze-up and breakup date determination in the Harbin, Heihe, and Raohe reaches. The framework extracts inter-station dependence from lagged correlations of temperature increments and combines correlation-based spatial weighting with annual, semiannual, and diurnal periodic encodings. It further employs an echo state network with an adaptive leaking rate and introduces horizon- and regime-conditioned closed-loop residual correction to reduce recursive error accumulation in both target- and auxiliary-station forecasts. Across the three scenarios, the 10-run evaluation yields an average MAE of 1.980 ± 0.193 °C, an RMSE of 2.544 ± 0.218 °C, and an $R^2$ of 0.9761 ± 0.0030. The corresponding Step-1 and Step-56 MAEs are 1.195 ± 0.167 °C and 2.040 ± 0.230 °C, respectively. The corresponding freeze-up and breakup date prediction errors across the three ice seasons from 2023–2024 to 2025–2026 are 1.33 and 1.22 days, respectively. These results indicate that, under the data conditions and reach-specific criteria considered in this study, the proposed framework can provide useful temperature-based information for river-ice monitoring, navigation management, and ice-season safety decisions.

Background: Improving container terminal efficiency requires a comprehensive understanding of the interactions between vessel, truck, and container operations, yet existing studies often analyzed these components separately. In Japanese container terminals, where digitalization initiatives are progressing, empirical evidence based on integrated operational data remains limited. Methods: This study empirically analyzes turnaround times for vessels, trucks, and containers at five major Japanese container terminals using a composite dataset that integrates terminal operating system data, automatic identification system data, and liner service information. Descriptive statistical analyses and regression models are applied to examine vessel berthing time, truck arrival patterns and turnaround time, container dwell time within terminals, and container round-trip time outside terminals. Results: The analysis reveals distinct temporal patterns in terminal operations, including systematic morning–afternoon asymmetries and differences across cargo flows. Truck turnaround times increase with vessel calls and vary by time of day, while container dwell times are strongly influenced by terminal policies such as free-time rules. Regression analyses indicate that turnaround times are primarily affected by terminal-controlled factors. Conclusions: These findings demonstrate the importance of synchronizing quayside and landside operations. The study contributes integrated empirical evidence to the port digitalization literature and provides actionable insights for enhancing container terminal efficiency.

Underwater marine debris detection is important for marine environmental monitoring, robotic inspection, and debris removal. However, reliable detection remains challenging because underwater images often suffer from low illumination, color distortion, turbidity, cluttered backgrounds, and weak boundaries. These factors reduce feature reliability and hinder accurate localization, especially for small, occluded, or low-visibility debris. To address these challenges, this paper proposes UADet, an adaptive detector for marine debris detection in degraded underwater scenes. UADet integrates two complementary components: Underwater Degradation-aware Feature Modulation (UDFM) and Visibility-aware Boundary Distribution Refinement (VBDR). UDFM extracts lightweight image-level degradation cues and modulates multi-scale features to improve robustness under varying underwater conditions. VBDR incorporates object scale and an appearance-based proxy for local visual difficulty into boundary distribution learning and matching cost, providing adaptive localization supervision for small objects and objects with weak visual evidence. Experiments are conducted on TrashCan and J-Litter, and UADet is compared with representative real-time detectors, including YOLOv8s, YOLOv10s, YOLOv11s, and RT-DETR. The results show that UADet achieves the best performance on both datasets, with 72.74% mAP@0.5, 81.36% precision, and 69.36% recall on TrashCan, and 48.02% mAP@0.5, 70.13% precision, and 55.61% recall on J-Litter. Compared with the strongest baseline, UADet improves mAP@0.5 by 3.12 percentage points on TrashCan and 4.83 percentage points on J-Litter. Ablation and qualitative analyses demonstrate that UDFM and VBDR provide complementary improvements. These results indicate that modeling underwater degradation and boundary uncertainty improves the robustness and reliability of marine debris detection in challenging underwater environments.

Abnormal ship behavior detection is important for maritime traffic surveillance, navigation safety, and risk prevention. However, existing methods often depend on handcrafted features or a single reconstruction or prediction signal, which limits their ability to detect both sustained trajectory abnormalities and abrupt vessel movement changes. This paper proposes a Dual-Error Fusion LSTM–Transformer framework, referred to as DEFLT, for AIS-based abnormal ship behavior detection. A motion-aware vessel representation was first constructed by combining the geographical position, speed over ground, course over ground, and their temporal variations. An LSTM autoencoder reconstructs historical trajectory windows, while a Transformer prediction module estimates subsequent vessel states. The standardized reconstruction and prediction errors are fused into a unified anomaly score to capture complementary evidence from historical trajectory inconsistency and unexpected future motion. Experiments were conducted using real-world AIS data collected during September 2019 from four representative Danish waters. The study considers four abnormal behaviors: speed anomalies, course anomalies, loitering, and route deviations. Compared with KNN, LOF, Isolation Forest, Random Forest, the LSTM-AE, and the Transformer, DEFLT achieves F1-scores of 0.96, 0.97, 0.88, and 0.93 across the four study areas. For type-specific detection, the Macro-F1 values range from 0.61 to 0.86, while Macro-Recall remains between 0.88 and 0.96. Friedman and post hoc Wilcoxon signed-rank tests further demonstrate that DEFLT provides a significant and consistent improvement over all baseline methods. These results verify the effectiveness of dual-error fusion for detecting heterogeneous abnormal ship behaviors from AIS trajectories. In operational settings, DEFLT can serve as an alert-prioritization tool for vessel traffic services and port authorities by directing attention to atypical trajectories that require timely review, thereby supporting safer and more resource-efficient maritime traffic coordination.