River and tidal dynamics regulate delta water levels, underpinning their ecological and socio-economic stability. Deltaic water level resilience, the capacity to recover from disturbances, is vital amid growing climate and human pressures, necessitating a robust method to quantify and monitor it. Here we define water level resilience as the rate of water level recovery post-hydrodynamic disturbances. We quantify resilience trends using autocorrelation and propose a high-dimensional resilience transmission method based on a vine Copula model to estimate the spatial propagation of resilience loss across subsystems. Applying this framework to the Pearl River Delta in China reveals spatial heterogeneity and inter-subsystem transmission of resilience loss driven by anthropogenic activities. Our approach can be extended to other major deltas, despite differing hydrodynamic and human impacts. The findings underscore the need for tailored management strategies addressing specific drivers and subsystem interactions, offering critical insights into the global delta preservation amid escalating environmental challenges.

This research proposes using a neural network to detect and identify the landmark points of the carapace of the Chinese mitten crab, with the aim of improving efficiency in observation, measurement, and statistics in breeding and sales. A 37-point localization framework was developed for the carapace, with the dataset augmented through random distortions, rotations, and occlusions to enhance generalization capability. Three types of convolutional neural network models were used to compare detection accuracy, generalization ability, and model power consumption, with different loss functions compared. The results showed that the Convolutional Neural Network (CNN) model based on the Differentiable Spatial to Numerical Transform (DSNT) module had the highest R² value of 0.9906 on the test set, followed by the CNN model based on the Gaussian heatmap at 0.9846. The DSNT-based CNN model exhibited optimal computational efficiency, particularly in power consumption metrics. This research demonstrates that the CNN model based on the DSNT module has great potential in detecting landmark points for the Chinese mitten crab, reducing manual workload in breeding observation and quality inspection, and improving efficiency.

This paper details the implementation of an integrated engineering framework for the real-time assessment of pose and size in fusiform fish, utilizing laser-camera technology. The design, comprising a camera and laser emitter, leverages laser triangulation for accurately measuring distances between key points, providing a reliable baseline for data comparison. Enhanced with the yolov7 model backbone, it includes detection and segmentation features, enabling precise image instance segmentation of fish and laser lines. The system’s dual-network structure, which combines fully connected regression and DSNT-MobileFaceNet networks, efficiently identifies six crucial landmarks on fish—an essential step for detailed pose analysis. This method facilitates the accurate determination of two-dimensional fish posture by analyzing the relative positions of these landmarks. A notable capability of this system is its ability to infer depth information from laser lines on the fish’s body, aiding in the accurate measurement of dimensions such as body length and depth. Empirical results demonstrate the system’s effectiveness, with high mean Average Precision (mAP) values for both object detection (0.9560 for fish, 0.8550 for laser lines) and segmentation (0.9740 for fish, 0.8420 for laser lines). The DSNT-MobileFaceNet network, in particular, shows excellent fitting accuracy with an $R^2$ value of 0.9170. The deep learning model achieves an average error rate of 7.75% in detecting fish data, markedly improving upon the baseline error rate of 14.70%. Overall, this study confirms the proposed system’s capability in accurately assessing fish pose and size. As a rigorous proof of concept validated in a controlled laboratory environment, this work establishes a foundational framework for non-invasive morphological monitoring, suggesting its future applicability in marine biology and aquaculture.

In order to solve the statistical problem of some targets when fishing vessels are operating, based on deep learning technology, this paper uses the improved YOLOv5s and SORT algorithms to achieve target statistics. First, YOLOv5s is fused with CBAM and SE attention mechanism modules, respectively, to reduce the interference of complex backgrounds and improve the model detection accuracy simultaneously. Comparing the three models, the target detection model with a better effect is selected. Secondly, through the threshold method, SORT algorithm combined with the detection line and DeepSORT three algorithms to achieve the count of some targets, respectively. The results show that the accuracies of YOLOv5s, YOLOv5s fused CBAM, and YOLOv5s fused SE are 97.2%, 84.8%, and 98.9%, respectively. Among them, the YOLOv5s fusion SE module has the best effect, which is 1.7% and 14.1% higher than the other two results. Among the three target statistics methods, the SORT algorithm combined with the detection line is the best, with an average count accuracy rate of 85.7%. The count accuracy rates of the three categories of Fish_basket, Fish_net, and Process_ship are 96.5%, 85.8%, and 75%, respectively, compared with the other two species have improved significantly. The research results can provide an informational reference for the automated counting of targets during fishing vessel operations.

Accurate characterization of wind-speed probability distributions underpins offshore wind resource assessment and ocean–atmosphere forcing estimates in coastal seas. Although the Weibull distribution is widely used, its limited shape flexibility often cannot represent the combination of a sharp central peak and a heavy right tail commonly observed across coastal–inland transition zones. Here, we introduce a physically consistent truncated proportional hazard logistic (TPHL) distribution. The model was obtained by conditioning the proportional hazard logistic family on nonnegative support, ensuring physical realism while retaining a closed-form, analytically tractable, 3-parameter structure with independent control of the typical wind level, the spread of the distribution core, and tail heaviness. We evaluated the TPHL distribution using long-term observations from 12 stations in the Hadley Centre Integrated Surface Database (HadISD) and a high-resolution wind dataset reconstructed from the fifth-generation atmospheric reanalysis (ERA5) produced by the European Centre for Medium-Range Weather Forecasts (ECMWF) for 2020 to 2024 across the Guangdong–Hong Kong–Macao Greater Bay Area. Benchmarked against the Weibull, gamma, and lognormal distributions, the TPHL distribution yielded superior goodness of fit with markedly lower Kolmogorov–Smirnov distances. Crucially, the TPHL distribution resolved the extreme value underestimation inherent in conventional models, reducing errors in wind-power density evaluation from approximately 30% to within 10% for most stations. The regionally fitted fields revealed a coherent inland–coastal contrast under typical conditions and variability. Overall, the TPHL distribution provides a unified distributional framework for wind-forcing characterization in complex land–sea environments, with direct value for coastal hazards and marine energy applications.