Search Results (421)

Amphibious assault vehicles serve as core equipment for coastal defense and amphibious operations worldwide, with irreplaceable strategic value. Featuring outstanding comprehensive performance, modern amphibious assault vehicles can maintain stable navigation under Sea States 3–4 and adapt to complex nearshore hydrological environments, emerging as the primary platform for mechanized landing operations of the Marine Corps. Cluster navigation is an inevitable tactical form in the operational application of amphibious assault vehicles. When multiple vehicles sail in formation, the wave-making and water pressure effects induced by individual vehicles generate prominent wave interference drag within the formation, which significantly impacts the overall navigation efficiency and stability. Based on the nearshore combat background of amphibious landing, this paper investigates different formation layouts of amphibious assault vehicle clusters to determine the optimal configuration for group navigation. First, a numerical simulation and a physical experiment are combined; a certain type of amphibious assault vehicle is taken as the prototype for 3D geometric modeling via SOLIDWORKS. Then, adopting the CFD numerical simulation method, with navigation speed and optimal inter-vehicle spacing fixed, variables including formation layout and number of vehicles are controlled to simulate the flow field characteristics and total resistance of different cluster formations in calm water. Meanwhile, 3D printing technology is applied to manufacture scaled-down models for towing tank tests. The experimental results are in good agreement with numerical simulations, revealing the fundamental hydrodynamic laws of formation navigation. Under optimal inter-vehicle spacing, the longitudinal tandem formation achieves the best drag-reduction effect, while the double-column staggered formation (diamond/V formation) can effectively suppress wave interference drag and improve the overall hydrodynamic performance and tactical coordination. The research provides a solid theoretical basis and data support for optimizing formation sailing strategies, enhancing cluster navigation stability and safety, and improving maritime maneuver efficiency. It is also of universal reference value for the tactical deployment of amphibious combat equipment globally. Full article

Monitoring complex river networks in Myanmar is challenging due to the high spatial heterogeneity and fragmentation of surface water bodies. Accurate identification of surface water resources is therefore essential for water resource management and for improving preparedness against climate change–induced hydrological hazards. To address this problem, this study proposes an adaptive interwoven deep learning–based segmentation framework that jointly utilizes multispectral reflectance information and topographic elevation data to enhance the extraction of fragmented water bodies. The framework is designed to coordinate feature interaction across spectral, spatial, and topographic dimensions by integrating channel-wise feature recalibration and attention-guided feature modulation within the encoding–decoding architecture. Experimental results demonstrate that the proposed method outperforms several traditional water index–based approaches, conventional machine learning algorithms and deep learning models. Across five independent training runs, the proposed framework achieves an average precision of 91.0%, recall of 93.5%, and F1 score of 92.3% (95% confidence interval: 91.8–93.0), demonstrating stable performance for fragmented water-body extraction. Cross-site experiments across three within-country study areas further demonstrate the spatial transferability and robustness of the proposed framework across diverse hydrological conditions within Myanmar. Overall, the proposed approach provides a reliable solution for fragmented water body extraction under heterogeneous hydrological conditions within Myanmar. Full article

Changes in river channel morphology under the influence of natural and anthropogenic factors pose a serious threat to the stability of aquatic ecosystems and water resource use, especially in regions with limited hydrological information. This study presents, for the first time, a three-dimensional numerical investigation of channel processes in the Talas River (Kazakhstan), employing the Volume of Fluid (VOF) method for free-surface flow simulation and a non-Newtonian model for sediment transport and riverbed morphodynamics. To verify the developed mathematical model, experimental data on the flow in the L-shaped channel and Earthfill dam break were used, which provided high reliability of the calculated results. The calculations showed a significant increase in the channel area in the studied section of the Talas River (from 41,334.92 m² to 56,890.17 m²) for the period from 2019 to 2024, mainly due to the intensification of the dynamics of currents and the formation of additional vortex zones with a diameter of 50 to 200 m. It was found that in places of local flow acceleration, water velocity increased up to 4.5 m/s, leading to bank erosion and channel widening, whereas after redistribution of channel flows, the maximum velocity decreased to 2.8 m/s, ensuring stabilisation of morphological changes. The results of the study underline the need for an integrated approach to river morphodynamics management using numerical modelling to predict channel changes, minimise flood risks and optimise the use of water resources. The presented computational approach can be adapted to analyse hydrodynamic processes in other poorly studied river systems, which significantly expands its scientific and practical value. Full article

River ice semantic segmentation is a crucial task that provides essential information for hydrological monitoring and infrastructure protection in cold regions. Previous works mainly focus on global long-range dependency modeling or local feature extraction, while the balance between computational efficiency and fine irregular-boundary preservation is often neglected. In this paper, we propose IceRWKV, an efficient semantic segmentation network for river ice based on the Receptance Weighted Key Value (RWKV). First, the RWKV sequence model is introduced into this task to break the quadratic complexity bottleneck, achieving high-precision global–local feature aggregation with low computation cost. Then, a novel Geometry-Direction Co-sensing Module (GDCM) is adopted to fit irregular ice contours and suppress background noise through an adaptive geometric correction and polarization feature-refinement strategy. Furthermore, Haar wavelet downsampling (HWD) is utilized to replace traditional downsampling operations, effectively mitigating feature aliasing and preserving high-frequency details. We conduct extensive experiments on the NWPU_YRCC_EX, NWPU_YRCC2, and Alberta River Ice Segmentation datasets. Comprehensive experimental results demonstrate that IceRWKV achieves state-of-the-art (SOTA) performance against 10 competing methods. Specifically, on the NWPU_YRCC_EX dataset, our method achieves a Mean Intersection over Union (mIoU) of 93.41% and an inference speed of 37.19 Frames Per Second (FPS) on NWPU_YRCC_EX, demonstrating a favorable trade-off between segmentation accuracy and computational efficiency. Full article

The lightning withstand performance of transmission lines is critically affected by grounding impulse characteristics, particularly for river-crossing towers where soil conditions are complex. This study develops a coupled seepage–electric field model to evaluate these characteristics under dynamic hydrological influences. A complex soil model is constructed integrating Bernoulli’s laminar flow equation with Richards’ equation for unsaturated seepage; long-term finite-element iterations simulate seepage dynamics, yielding distributed soil conductivity parameters that vary with river flow velocity, water depth, and impermeable layers. These parameters are then coupled with an electroquasistatic Maxwell framework to model impulse current dispersion. Validation against experimental data confirms the model’s accuracy. Results show that seepage increases moisture and lowers resistivity. Increasing flow from static to 10 m/s reduces riverbed pressure from 5.61 × 10⁴ Pa to 1.86 × 10⁴ Pa, shifting the 0 Pa isobar downward by 5.1 m, weakening seepage and raising impulse resistance. A shallower impermeable layer deflects seepage laterally, reducing nearby resistivity. Raising water depth from 5 m to 10 m increases pressure from 1.96 × 10⁴ Pa to 5.61 × 10⁴ Pa, enhancing seepage and lowering resistivity. These findings indicate that grounding design must holistically account for flow velocity, water depth, and subsurface barriers to ensure reliable lightning current dissipation and stable grid operation. Full article

Accurate rainfall input is critical for reliable flood simulation, particularly in semi-arid watersheds with pronounced spatiotemporal precipitation heterogeneity. This study developed a radar–gauge rainfall fusion framework to improve HEC-HMS (Hydrologic Engineering Center–Hydrologic Modeling System) model performance in the Liulin experimental watershed, Xingtai City, Hebei Province. Radar quantitative precipitation estimation (QPE) was generated via a dynamically optimized Z-I relationship, then fused with gauge observations using three methods—Geographical Differential Analysis (GDA), Conditional Merging (CM), and Random Forest (RF). The fused products drove a calibrated HEC-HMS model, evaluated over five representative flood events. All three methods corrected radar QPE underestimation. Under independent cross-validation, GDA and CM achieved comparable point-scale accuracy (CC ≈ 0.81, RMSE ≈ 5.7 mm), while RF showed lower generalization (CC ≈ 0.48, RMSE ≈ 8.7 mm) due to overfitting. In flood simulations, GDA performed most robustly, followed by RF and CM, all surpassing single-source inputs. Notably, CM’s higher statistical accuracy did not translate into better flood performance, indicating that optimal statistical fidelity does not guarantee optimal hydrological results. Peak discharge deviations persisted for short-duration intense storms and long-duration uneven rainfall events. This study confirms that radar–gauge fusion enhances rainfall input quality and provides a reliable approach for improving flood forecasting. Full article

Accurate prediction of extreme runoff events is crucial for flood control and disaster mitigation and water resource risk management. In recent years, deep learning has emerged as a key method for runoff simulation. However, its training strategy applies uniform optimization across all time steps, which encourages the model to fit normal hydrological processes while reducing its sensitivity to extreme runoff events. To address this issue, this study introduces Selective Learning and proposes a runoff prediction model with a flood peak protection mechanism, namely PPSL-DI-LSTM. In this framework, sample importance is dynamically regulated through uncertainty filtering, anomaly filtering, and flood peak protection, thereby reducing the influence of unstable or non-generalizable samples while enhancing the learning of high-flow processes. Based on this strategy, a dynamic weighted loss function, termed Selective Weighted RMSE (SW-RMSE), is designed to optimize the training process. Experimental results on the CAMELS dataset show that, compared to the baseline DI-LSTM model, the proposed model achieves improvements of approximately 4% and 6% in median Nash-Sutcliffe Efficiency (NSE) and median Kring-Gupta Efficiency (KGE), respectively. Simultaneously, the Flow Duration Curve High Flow Variability (FHV) was significantly reduced by approximately 84% and the overall percent bias (PBIAS) improved from −3.80% to 1.15%. Additionally, when integrated into a standard LSTM, PPSL maintains comparable median NSE while improving median KGE (from 0.751 to 0.778) and substantially reducing FHV (from −12.86% to −3.34%). Our method overcomes the limitations of uniformly treating all training samples in complex hydrological simulations, and has strong potential to substantially improve the accuracy and reliability of future runoff modeling studies. Full article

Disentangling the effects of climate variability and forest management on catchment hydrology remains a major challenge in temperate plantation landscapes. We analyzed 21 hydrological years (1997/98–2017/18) of precipitation and runoff records from two experimental catchments in the Chilean Coastal Range with contrasting silvicultural histories to characterize changes in low-flow behavior. Hydroclimatic and low-flow indices were evaluated using the monotonic trends test, Sen’s slope estimates, change point detection, and standardized inter-catchment anomaly differences. Annual and seasonal precipitation indices, rainfall frequency, and maximum dry-spell duration showed no significant monotonic trends, whereas maximum 5-day precipitation declined at LP. The two catchments nevertheless exhibited divergent low-flow trajectories. Los Pinos, managed through partial harvesting and thinning within a forest mosaic, showed decreasing normalized low-flow availability and longer low-flow exposure during the latter part of the record. La Reina, clearcut in 1999/2000 and subsequently reforested, showed a progressive increase in low-flow magnitude and normalized low-flow availability, together with declining flow variability and fewer below-threshold events. Standardized inter-catchment comparisons confirmed a temporal divergence in low-flow behavior. They also revealed a concurrent shift in inter-catchment precipitation anomalies. These results indicate contrasting long-term hydrological trajectories that are consistent with differences in forest management histories; however, the non-paired study design, limited pre-harvest observations at La Reina, and differential precipitation forcing preclude formal attribution to silvicultural effects alone. This study highlights the value of long-term experimental catchments for evaluating low-flow dynamics under interacting climatic and land-management influences. Full article

Accurate prediction of reservoir discharge and sediment load is essential for optimizing reservoir operations and mitigating downstream flood risks. In the middle reaches of the Yellow River, water–sediment evolution exhibits significant nonlinearity and temporal dependency. However, existing models often neglect the dynamic coupling between water–sediment evolution and storage–discharge states and lack hydrological prior constraints. To improve performance under complex conditions, this study developed models for the Sanmenxia and Xiaolangdi reservoirs based on historical hydrological data from 2002 to 2022, with features selected via Pearson correlation. By incorporating variables such as reservoir capacity, we evaluated four deep learning architectures: CNN, LSTM, CNN-LSTM, and TCN. Furthermore, a Genetic Algorithm (GA) was employed to optimize the hyperparameters, utilizing a custom fitness function that integrates Nash–Sutcliffe Efficiency (NSE) with non-negative boundary constraints to ensure the simultaneous optimization of predictive accuracy and hydrological consistency. Experimental results demonstrate that the CNN-LSTM model achieved superior performance. Specifically, the test set NSE values for discharge and sediment reached 0.927 and 0.821 for Sanmenxia, and 0.949 and 0.843 for Xiaolangdi, respectively. In outflow prediction, the feature weights of inflow and reservoir capacity for Sanmenxia exceed 45% and 35%, while for Xiaolangdi, those of upstream and local inflow exceed 40% and 55%, respectively. In sediment prediction, almost all feature weights exceed 15%. This research provides a valuable reference for intelligent reservoir management in the Yellow River, advancing deep learning applications in complex hydrological systems. Full article

The aim of this study is to develop an integrated methodological framework for assessing flood susceptibility, agricultural land vulnerability, and landscape structure using Geographic Information Systems (GIS) and the Analytic Hierarchy Process (AHP), thereby contributing to the integrated assessment of these components at the scale of an experimental micro-watershed. The research was based on a 2.5 m resolution Digital Elevation Model (DEM) and detailed data obtained through field investigations. Flood susceptibility was modeled by integrating geomorphological, hydrological, pedological, and land-use factors. The results highlight the predominance of the moderate susceptibility class (64.53%), while the high and very high susceptibility classes account for 14.22% of the watershed area and are concentrated in the vicinity of the drainage network and in low-lying sectors. The assessment of land vulnerability revealed differences among land-use categories. Hayfields encompass the largest area exposed to high and very high flood susceptibility classes, whereas forested areas are predominantly associated with low and moderate susceptibility classes. The landscape structure analysis revealed very strong correlations between the exposed area and the landscape metrics used to characterize landscape structure. The proposed integrated methodological framework provides a transferable approach for the simultaneous assessment of flood susceptibility, agricultural land vulnerability, and landscape structure in micro-watersheds with similar characteristics. Full article

Flood forecasting is crucial for predicting floods and facilitating timely evacuations. Machine learning and deep learning algorithms, such as artificial neural networks (ANN), recurrent neural networks (RNN), and Transformer models, have shown significant success in time series prediction tasks. Recently, Diffusion models, which add noise to training data and then reverse the process to recover the data, have gained popularity in data generation. In flood prediction, however, the limited availability of hydrological data from reservoirs often hinders the training of deep learning models. This study proposes a novel approach by combining the Diffusion model with the Transformer model to address the issue of insufficient data. The Diffusion model is used for data augmentation, while the Transformer model is employed for flood flow prediction. For multivariate input datasets, we introduce a Convolutional Block Attention Module (CBAM) into the Transformer. It can adaptively learn the importance weights of different input variables. Additionally, considering the rich combinations of real-world time series—such as temporal trends, periodicity, and local specificities—which are often disrupted by the gradual addition of noise during the Diffusion process, we incorporate a time trend module into the Diffusion model. This enhancement allows the Diffusion model to better extract the temporal characteristics of the original data during generation, producing new data that retains more of the original information and improves the effectiveness of data augmentation. By combining the improved Diffusion and Transformer models, leveraging the powerful generative capability of the enhanced Diffusion model and the better predictive ability of the enhanced Transformer model, the prediction performance is significantly enhanced. Experimental results demonstrate that, in most cases, our proposed model achieves better prediction performance compared to ANN, LSTM, Transformer models and the other flood prediction methods. Full article

Existing cache replacement strategies in large-scale spatiotemporal data systems struggle to cope with complex and dynamic access patterns characterized by long-tail distributions and periodic behaviors. Traditional heuristic-based methods such as Least Recently Used (LRU) and Least Frequently Used (LFU) frequently fail to generalize across varying workloads, while recent learning-based approaches are limited by their reliance on hand-crafted features or short-term dependencies. In this paper, we propose a cache replacement framework named CorrelaCache, which integrates imitation learning with a temporal autocorrelation mechanism to capture both short-term and long-range periodic access patterns. By modeling the replacement task as a Markov Decision Process (MDP) and using the Belady optimal policy as the supervision signal, our method adopts Long Short-Term Memory (LSTM) networks for sequential encoding and employs Fast Fourier Transform (FFT)-based autocorrelation to detect and align periodic phases in access history. We further incorporate a joint prediction layer and a hybrid loss function that combines ranking loss and reuse distance prediction loss, and mitigate distributional shift during training via the Dataset Aggregation (DAgger) algorithm. Experimental results on five public meteorological datasets with generated hydrological access traces show that CorrelaCache outperforms representative baselines in the evaluated workloads. Full article

Ecological porous concrete (EPC) offers both structural performance and ecosystem services, yet an inherent contradiction exists between the ecological benefits of high porosity and mechanical performance. Traditional design methods focusing solely on macro-scale porosity fail to achieve synergistic optimization. This review comprehensively synthesizes the intrinsic correlations between EPC’s multi-scale pore structures and key properties from micro-, meso-, and macro-scale perspectives, drawing upon representative studies across experimental, numerical, and theoretical approaches. The microscale reveals interfacial transition zone bonding, capillary pore effects, and alkalinity regulation for vegetation compatibility. The mesoscale clarifies the control of effective porosity, tortuosity, and pore throats on fluid transport and root penetration. The macro-scale analyzes skeletal pore support for plant growth, hydrology, and slope stability. A cross-scale collaborative design approach is proposed, featuring microscopic reinforcement, mesoscopic continuity, and macroscopic moderation. This paper provides theoretical support for EPC’s transition from empirical to precision design, promoting low-carbon and large-scale applications in revetments, Sponge Cities, and slope restoration. Full article

Plant communities in reservoir drawdown zones experience highly altered hydrological regimes, and responses of locally dominant species shape the biodiversity and restoration trajectories of these artificial wetlands. The water-level fluctuation zone (WLFZ) of the Three Gorges Reservoir (TGR) is exposed to alternating flooding and drought, which strongly constrains both its vegetation and the biodiversity that depends on it. Cynodon dactylon dominates the herbaceous cover of the TGR WLFZ, but evidence on its stress responses remains fragmented across single-site studies. Following a PRISMA 2020 literature search and screening procedure, we synthesized 169 effect sizes from 12 qualifying experimental studies, covering biomass and morphological traits, photosynthetic gas-exchange parameters, chlorophyll content, and oxidative-stress indicators. Effect sizes were calculated as natural log response ratios (lnRR) and pooled with random-effects models; shallow and deep flooding were compared using subgroup analyses with bootstrap 95% confidence intervals. Flooding effects varied with water depth. Shallow flooding increased total biomass (+47.2%), whereas deep flooding reduced plant height (−46.5%) and root length (−22.3%). Plant height showed significant between-group heterogeneity (Q_between = 5.60, p = 0.045), indicating sensitivity to submergence depth. Flooding also increased malondialdehyde content (MDA) by 31.7%, whereas peroxidase activity (POD), superoxide dismutase activity (SOD), and photosynthetic gas-exchange parameters showed no consistent responses. Drought effects on total biomass, plant height, and total chlorophyll were non-significant, although inference was limited by a few drought-related entries. Deep flooding, therefore, appears to be a stronger constraint than drought for Cynodon dactylon in the TGR WLFZ, mainly through morphological suppression and increased oxidative damage. Given the dominant role of this species in the herbaceous layer, its depth-dependent decline is relevant both for biodiversity conservation in this artificial wetland and for elevation-based restoration planning. Full article

Floods remain among the most devastating natural disasters globally, disproportionately impacting low-resource regions where real-time flood forecasting is constrained by limited computational infrastructure and the scarcity of fine-resolution predictive models. Although state-of-the-art global climate models achieve high predictive accuracy, their scale and computational complexity restrict their applicability in localized and resource-constrained settings. This study proposes a deep climate model distillation framework that transfers knowledge from a high-capacity Fourier Neural Operator (FNO)-based global climate model inspired by FourCastNet into lightweight, regionally adaptive student networks suitable for edge deployment. The framework combines climate variables, satellite observations, and hydrological measurements to improve localized flood prediction. Knowledge transfer is achieved through a multi-objective distillation strategy that combines supervised learning, soft-target alignment, and intermediate feature matching. Experimental evaluation across multiple flood-prone regions in Sub-Saharan Africa and South Asia shows that the distilled student model achieves an average classification accuracy of 0.89, an AUC of 0.91, and an F1-score of 0.88, retaining approximately 96.7% of the teacher model’s predictive performance. In continuous discharge estimation, the model attains a mean absolute error of 0.17, RMSE of 0.24, and an R² score of 0.85. The proposed distillation approach yields an 8× reduction in inference latency and over a 20× reduction in model size, enabling real-time execution on low-power edge devices such as the Raspberry Pi 4 and NVIDIA Jetson Nano. The student model further demonstrates robust regional and temporal generalization, with limited performance degradation in unseen geographic areas and during extreme flood years. Full article