[Objective] This study centers on the dike fields of the spur dike group in the Yangtze River Estuary, a typical tidal estuary where complex water-sediment dynamics and diverse dike layouts jointly shape deposition processes. Its core objectives are twofold: first, to unravel the coupling mechanism through which dynamic factors (e.g., runoff, tides) and geomorphic parameters (e.g., dike spacing, initial water depth) jointly regulate sediment deposition intensity in tidal estuarine dike fields; second, to develop a reliable empirical formula for predicting such deposition intensity. By addressing the gap in existing research—where the integrated effects of dynamic and geomorphic factors are often overlooked—this study aims to provide robust theoretical support for optimizing the design of spur dike groups and enhancing the accuracy of deposition forecasting in the Yangtze River Estuary and analogous tidal estuarine systems worldwide. [Methods] The dike fields of the spur dike group in the north passage of the Yangtze River Estuary, a key area of the Yangtze River Estuary Deepwater Channel Regulation Project, were selected as the research focus. Long-term, systematic measured data were analyzed, including dike field topographic surveys, hydrological observations, and sediment monitoring records. Correlation analysis was first performed to examine how deposition intensity relates to key dynamic factors (upstream runoff from the Datong Hydrological Station, suspended sediment concentration, offshore tidal range, suspended sediment particle size) and critical geomorphic parameters (relative spacing of spur dikes, initial water depth of dike fields, spur dike length, dike field depth). Using dimensional analysis and the Buckingham π theorem, a comprehensive dynamic parameter was constructed by integrating the four dynamic factors, synthesizing their combined influence on water-sediment transport and deposition. Simultaneously, a set of geomorphic parameters was established, incorporating spur dike spacing, length, and dike field depth to quantify the impact of spur dike group layout and dike field topographic features on local flow patterns and sediment trapping. A power function model was then used to quantify the coupling relationship between the comprehensive dynamic parameter and geomorphic parameters, and an empirical formula for deposition intensity was derived. Finally, the formula was validated using measured data from representative dike fields, including those unaffected by subsequent engineering and those influenced by phased projects. [Results] 1) As the Yangtze River Estuary Deepwater Channel Regulation Project advanced through three phases, total sediment deposition in dike fields increased significantly (from 15.48×106 m3 in Phase I to 128.01×106 m3 in Phase II), confirming the spur dike group’s strong sediment-trapping effect. 2) Deposition intensity was positively correlated with runoff (higher runoff carries more sediment to dike fields) and sediment concentration (more available sediment for deposition), but negatively correlated with tidal range (larger tidal range strengthens ebb currents, enhancing offshore sediment transport) and sediment particle size (coarser particles settle before reaching dike fields or are easily resuspended by strong flows). 3) Among geomorphic parameters, initial dike field water depth showed a strong positive linear correlation with deposition intensity (deeper water provides more deposition space and reduces flow velocity, favoring sediment settlement), while spur dike relative spacing had weak correlation (R2=0.44), due to interactions with factors like flow blockage (too small spacing) or uneven energy distribution (too large spacing). 4) The comprehensive dynamic parameter correlated highly (R2=0.94) with annual deposition in undisturbed dike fields (TS1, TS2), effectively capturing dynamic drivers of deposition; geomorphic parameters correlated strongly (R2=0.96) with initial deposition, clearly distinguishing differences between dike fields in the same spur dike group. 5) The empirical formula showed excellent agreement with measured data: it matched well with the measured deposition intensity of TS1, TS2, and TS8 (used for fitting analysis) and effectively reflected the deposition intensity of TN7, TN8, and TN9 (used for validation in the second-phase project). Even for dike fields affected by phased engineering or new structures (e.g., a 21 km sediment barrier), the formula still successfully captured the overall deposition trend. [Conclusion] This study makes three key contributions: it innovatively integrates dynamic factors and geomorphic parameters into a unified analytical framework for Yangtze River Estuary spur dike group dike fields, overcoming the limitations of previous single-factor research; the constructed comprehensive dynamic parameter and geomorphic parameters effectively quantify the combined effects of water-sediment dynamics and dike layout/topography on deposition, making complex processes interpretable; the empirical formula, with high applicability and accuracy, offers a reliable tool for tidal estuarine dike field deposition prediction.
[Objective] The Jingjiang-Dongting Lake region, a critical river-lake system in the middle reaches of the Yangtze River, is crucial for flood mitigation, sediment regulation, and ecological conservation. This study aims to develop and apply an enhanced 2D water-sediment model incorporating a morphological acceleration factor (MF) to efficiently predict the bed deformation and sediment redistribution over a 30-year period, thereby providing a scientifically sound and computationally feasible tool for long-term morphodynamic prediction and sustainable management of large river-lake systems. [Methods] A 2D depth-averaged water-sediment model was established to solve shallow water equations and multi-fraction suspended sediment transport equations coupled with a bed deformation module. The governing equations were discretized using an unstructured finite volume method, with advection terms solved by the Euler-Lagrange method (ELM) to enhance numerical stability. To overcome the computational bottleneck of long-term simulations, the MF was introduced, linearly scaling the bed evolution per hydrodynamic time step to accelerate morphodynamic simulations while maintaining physical realism. The model domain covered the lower Jingjiang reach and the Dongting Lake and was discretized into unstructured cells with refined resolution along the main channel. Sediment was grouped into three size classes for suspended load and four for bed material, with gradation initialized from field surveys from 2003 to 2012. MF was determined at 7, 15, and 24 for sensitivity analysis. [Results] (1) The MF markedly improved computational performance. With MF values of 15 and 24, simulation time decreased to 42% and 30% of that for MF=7, corresponding to speed-up factors of 2.38 and 3.33, respectively. The spatially distributed erosion-deposition patterns remained consistent across different MF values, confirming the robustness of the approach. The total sediment deposition in Dongting Lake varied by less than 5% across scenarios, while the total erosion volume along the main stem exhibited higher sensitivity, with a maximum deviation of 9.1% between MF=24 and MF=7. These deviations were primarily localized and did not alter the long-term trends or magnitudes. (2) The mainstream of the Jingjiang River experienced sustained incision, with a total scour volume of 462 million m3 and an average bed incision of 1.86 m over 30 years. Local deposition occurred along convex banks due to curvature-induced secondary flows. Dongting Lake exhibited net deposition of 276 million m3, with an average siltation thickness of 0.09 m. Notably, the annual deposition rate in the lake decreased significantly over time—from 20 million m3 to about 6 million m3—representing an approximate 70% reduction and indicating a gradual approach toward a new morphodynamic equilibrium. Significant spatial variability in sediment redistribution was observed. The Jingjiang mainstream was dominated by scour, particularly in the deep channel, while point bars developed in its meandering segments. Within Dongting Lake, distinct patterns emerged. The western Dongting area was near equilibrium with no clear trend, the southern Dongting experienced significant deposition along floodways from the Three Outlets with thicknesses up to 4 meters, and the eastern Dongting exhibited complex patterns with both deep scour pits infilled by sediment and a depositional bar at the Zhuzikou inlet. Furthermore, the outlet channel in the lake-river confluence zone experienced upstream erosion and downstream deposition, gradually flattening the longitudinal slope. Regarding model validation, the simulated results closely aligned with previous studies and field observations. The deviation in total deposition was 17.0% compared to a 1D model over 30 years, and the deviation was approximately 14.2% at 10 and 20 years compared to an earlier 2D lake model. Notably, the predicted average annual deposition of 89.3 million m3 from 2011 to 2020 closely matched the measured data, further supporting the model’s reliability. [Conclusion] This study demonstrates that integrating a morphological acceleration factor into a 2D water-sediment model enables efficient and accurate simulation of decadal-scale morphodynamics in large river-lake systems. The MF method decreases computational time for a 30-year simulation by about 60%-70% while maintaining prediction errors within 5% for total lake deposition. The Jingjiang reach is projected to undergo continued incision, while Dongting Lake will experience slight net deposition at a strongly declining rate, indicating adjustment toward a new dynamic equilibrium. Spatially heterogeneous erosion-deposition patterns highlight the need for targeted management strategies.
[Objective] Illegal sand mining in river channels has caused riverbed incision and led to eco-environmental problems such as permafrost degradation, bank slope instability, and destruction of aquatic ecosystems, threatening the ecological health of plateau rivers and lakes in the Xizang Autonomous Region. This study aims to analyze the current situation and problems of river sand mining management in Xizang, propose management and protection measures, and provide references for promoting the standardization of river sand mining management in the Xizang Autonomous Region. [Methods] Methods including investigation, data statistics, empirical analysis, qualitative analysis, and comprehensive analysis were employed to systematically review the current status of river sand mining management in the Xizang Autonomous Region and analyze problems in the construction of laws and regulations, river sand mining planning, preparation of annual sand mining implementation plans, and management of engineering sand mining. [Results] River sand mining management in the Xizang Autonomous Region faces problems such as incomplete laws and regulations, insufficient operability and specificity of some institutional requirements, and the presence of blind spots in management. In some areas, river sand mining plans and annual sand mining implementation plans are not prepared according to relevant technical guidelines or standards, resulting in insufficient scientific basis and compliance of the plans or annual implementation plans. The mechanism for defining engineering sand mining requires further improvement. In daily supervision, the management initiative of water administrative departments in some areas needs to be strengthened, with limited regulatory measures and relatively weak technical capacity. These problems affect the efficiency of management and supervision of river sand mining. [Conclusion] This study proposes the following countermeasures and recommendations. (1) Efforts in legislative research, local regulations, and legal constraints should be strengthened in full consideration of the management situation of river sand mining in the Xizang Autonomous Region. (2) By refining the management requirements for river sand mining and formulating standards for illegal activities and penalties, scientific river sand mining plans should be developed, taking into account the characteristics of river sections, current status of sand mining, and ecological management needs of rivers and lakes in the planning period for each planned reach. In light of the widespread distribution of glaciers and permafrost, long vegetation recovery cycle, and chain reactions like riverbed incision and shoreline collapse triggered by sand mining activities that threaten water conservation functions, annual river sand mining plans and technical standards for the comprehensive utilization of dredged sand should be formulated to strengthen the top-level design of river sand mining management. (3) For special areas such as uninhabited areas, differentiated supervision plans should be developed. The river chief system platform should be fully utilized to strengthen joint law enforcement operations among multiple departments, carry out joint law enforcement at the junctions of administrative regions, enhance targeted governance in areas with weak regulatory capacity, and strengthen on-site supervision for permitted sand mining sites by scale and region, thereby improving daily supervision mechanisms. (4) A complete management system for river sand and gravel mining and transportation should be established, an integrated monitoring and perception system for river sand mining should be gradually developed, and the construction of intelligent supervision should be promoted. These recommendations can provide theoretical support and practical reference for the supervision of river sand mining with plateau ecological characteristics.
[Objective] Poyang Lake is the largest lake connecting to the Yangtze River. A complex river-lake relationship is observed between the Yangtze River and Poyang Lake, manifested as hydrodynamic processes such as water level uplift and flow field changes. In recent years, extreme drought in Poyang Lake and floods in the Yangtze River have become increasingly frequent amid repeated flood-drought alternations, resulting in significant changes in the river-lake relationship. This study aims to clarify this relationship for the Yangtze-Poyang confluence system. [Methods] To simulate the natural flow state and typical inflow conditions, we established a large-scale generalized physical model of about 80 meters in length and 45 meters in width. The model consisted of the Yangtze River section (from Jiujiang to Balijiang) and the waterway of Poyang Lake (from Xingzi to Hukou). A total of nine flow conditions representing the combined hydrological characteristics of the Yangtze River and Poyang Lake were selected for the experiments. The cases covered typical natural flow conditions such as floods and droughts in the Yangtze River and Poyang Lake, as well as confluence ratios ranging from 1 to 60 at the Yangtze River-Poyang Lake confluence. [Results] (1) When the confluence ratio of the Yangtze River and Poyang Lake was 10, the jacking coefficient α was greater than 1, and the energy coefficient Fe was less than 0. When the flow discharge of the Yangtze River further increased and the confluence ratio reached 60, α and Fe reached their maximum and minimum values. When the confluence ratio was about 0.5, the river-lake relationship exhibited opposite effects. (2) The evolution of the river-lake relationship was closely related to the flow conditions of the Yangtze River and Poyang Lake. The jacking coefficient α was nonlinearly positively correlated with the confluence ratio of the two flows, and weakly linearly positively correlated with the velocity ratio. The energy coefficient Fe showed a nonlinear negative correlation with the confluence ratio, and a weak linear negative correlation with the velocity ratio. (3) When jacking effect occurred, the water level difference between Xingzi and Hukou in Poyang Lake decreased, and the flow velocity distribution tended to be uniform. At the confluence, the Yangtze River water flowed upward from the middle and lower parts and interacted with the Poyang Lake water. As the Yangtze River discharge further increased, the water level drop decreased. There was also a backflow on the right side of the confluence, causing Yangtze River water to flow back into Poyang Lake. When the flow rate of Poyang Lake reached 25 000 m3/s, there was a significant increase in the water level drop and cross-sectional flow velocity. The water flow of Poyang Lake mainly entered the Yangtze River near the water surface. [Conclusion] When the confluence ratio exceeds 45, it is recommended to regulate the outflow of reservoirs on the Yangtze River in advance to reduce the water level and mitigate backflow. Under extreme low water conditions where the confluence ratio is less than 0.5, a water level gradient threshold (0.45‰) is used as an indicator. The scheduling of water conservancy projects is suggested to be combined with these measures to increase the water replenishment flow and reduce the water level gradient in the Poyang Lake waterway, thereby alleviating the water level reduction effect.
[Objective] The simplified structures of traditional hydrological models often limit their adaptability under complex hydroclimatic and anthropogenic conditions. Meanwhile, data-driven models such as deep learning frameworks typically lack explicit physical interpretability. To address these challenges, this study compares the performance of three representative runoff prediction methods: two physically based models—the Xinanjiang (XAJ) model and Soil & Water Assessment Tool (SWAT) model—and one data-driven model, the Long Short-Term Memory (LSTM) model. The Xiangjiang River Basin, a major tributary of the Yangtze River in southern China, is selected as the study area. This study aims to evaluate model performance and adaptability across multiple temporal scales. [Methods] The dataset included daily runoff records from 1971 to 2020 at the Xiangtan hydrological station, along with concurrent meteorological observations from 32 meteorological stations. Land use data with a spatial resolution of 1 km × 1 km were obtained from the Chinese Academy of Sciences, and soil property data were derived from the Harmonized World Soil Database (HWSD). The employed XAJ model estimated evaporation from three soil layers, routed surface runoff using the unit hydrograph method, and modeled interflow and groundwater components through the linear reservoir approach. The SWAT model divided the river basin into 13 subbasins and 192 hydrological response units (HRU) using high-resolution spatial inputs. For both models, the SUFI-II algorithm was employed for parameter calibration based on the Nash-Sutcliffe Efficiency (NSE) coefficient and the total water balance error. The LSTM model was trained using climate indices as inputs under different strategies (areal averages and multi-station series). The input indices included precipitation, temperature, and relative humidity. To evaluate the effects of temporal dependencies, lag times from 0 to 5 days were tested. All three models were calibrated using data from 1971 to 2010 (with the first two years as a warm-up period) and were validated over 2011-2020. [Results] (1) For daily-scale prediction, the LSTM model achieved the highest accuracy, with NSE values reaching 0.99 during calibration and 0.87 during validation when using multi-station meteorological inputs. Incorporating spatially distributed meteorological inputs significantly improved LSTM performance compared to using areal averages, highlighting the importance of spatial heterogeneity in data-driven hydrological forecasting. The XAJ model performed robustly (NSE>0.76), especially during flood seasons, but tended to overestimate dry-season flows and underestimate flood peaks. The SWAT model (NSE≈0.6) reproduced the overall hydrograph patterns but showed systematic biases similar to those of the XAJ model. (2) At the monthly scale, the performance of the SWAT model improved significantly (NSE=0.92 during calibration, 0.83 during validation), while LSTM accuracy declined (NSE = 0.86 during validation). The reduced training sample size (480 months) likely caused overfitting in the LSTM model and limited its generalization ability. Incorporating temperature and humidity as input features enhanced the stability of the LSTM model, indicating that variables related to evapotranspiration became more influential at coarser temporal resolutions. (3) Both hydrological models showed distinct seasonal bias patterns: overestimation during dry seasons and underestimation during wet seasons. The main reason was the fixed parameterization that failed to represent temporal variability in infiltration and storage processes. In contrast, the flexibility of the LSTM model enabled better adaptation, though at the cost of physical interpretability. The observed discrepancies also reflected the impact of nine cascade hydropower projects along the mainstream of the Xiangjiang River, which regulated flow seasonality but were not explicitly modeled in the hydrological frameworks used in this study. [Conclusion] In summary, this study systematically evaluates the performance of the XAJ, SWAT, and LSTM models for runoff forecasting in the Xiangjiang River Basin at daily and monthly scales. The results demonstrate that the LSTM model achieves the highest forecasting accuracy and computational efficiency, particularly at the daily scale. In contrast, the XAJ model is more reliable during flood seasons, and the SWAT model is more suitable for representing the long-term spatiotemporal variability of hydrological processes. Although the data-driven LSTM model lacks explicit physical mechanisms, it offers substantial advantages in adaptability and predictive precision, whereas physically based models maintain interpretability and stability under limited data conditions. The innovation of this study lies in bridging process-based and deep learning methods under unified experimental conditions, emphasizing the potential of hybrid modeling frameworks that integrate physical constraints with deep neural architectures to enhance both accuracy and interpretability in future runoff forecasting applications.
[Objective] This study aims to analyze the changes in water resources in the middle and lower reaches of the Yangtze River and to investigate the impact of El Niño events on regional floods. [Methods] GRACE gravity satellite data from 2003 to 2022, released by various research institutions, were used to derive the Terrestrial Water Storage Anomaly (TWSA) for the study area. Based on correlation coefficients and cross-correlation analysis, the CSR MASCON TWSA data series exhibiting strong correlations with the indices of both Eastern Pacific (EP) and Central Pacific (CP) El Niño events was selected. Wavelet analysis, Empirical Orthogonal Function (EOF) analysis, and the Flood Potential Index (FPI) were employed to investigate the influence of the two types of El Niño events on regional TWSA and to analyze their relationship with flood risk in the study area. [Results] The results were as follows: (1) the highest correlation between TWSA and the EP El Niño event was found at a time lag of 6 months, with a correlation coefficient reaching 0.630. TWSA peaks showed a positive response to EP El Niño events, while the response to CP El Niño events was unstable. (2) Cross wavelet transform revealed common resonance periods between TWSA and both types of El Niño events, and the impact of the EP El Niño event on water resource changes in the middle and lower reaches of the Yangtze River was found to be more significant. The EOF analysis showed that the southern part of the study area was susceptible to the influence of both El Niño types. (3) The spatial distribution of the grid-based Flood Potential Index showed a higher flood risk in the southern part of the study area following the occurrence of both El Niño types. The flood risk corresponding to EP El Niño events was greater, with high-risk areas concentrated at the junction of the Dongting Lake and Poyang Lake sub-basins. In contrast, the flood risk distribution corresponding to CP El Niño events was more dispersed. [Conclusion] The results of this study, based on wavelet analysis, EOF analysis, and the Flood Potential Index, show that flood risk in the middle and lower reaches of the Yangtze River is closely related to El Niño events. These findings contribute to the prediction and prevention of floods in the region.
[Objective] To investigate the impact of the completion and operation of the south-side drainage project of the Qingcaosha Reservoir on the water quality of the Changxing Island river network, this study quantitatively assesses the improvement effects of the project’s drainage on the river network’s water quality under different operating conditions. [Methods] Based on detailed fundamental data, a one-dimensional hydrodynamic and water quality model for Changxing Island was constructed, calibrated and validated using measured data. The improvement effects of the south-side drainage project on the river network water quality were quantified under different operating conditions; furthermore, a water resource scheduling approach for the Changxing Island river network was proposed and optimized under the planned scale conditions through multi-scheme comparison. [Results] The model simulation analysis showed that when the Changxing Island pump-gate system operated under routine scheduling, the project operation could increase the proportion of river length with a permanganate index of Class Ⅲ and above from 25% to 35%-40%, and increase the proportion of river length with total phosphorus of Class Ⅱ and above from 72% to 77%-82%. The west-to-east drainage scheme demonstrated a significantly better improvement effect on total phosphorus in the Changxing Island river network than the routine scheduling scheme. Under the maximum flow condition of the project, this scheme could increase the proportion of river length with total phosphorus of Class Ⅱ and above to 95%. [Conclusion] The completion and operation of the south-side drainage project of the Qingcaosha Reservoir can effectively improve the water environment quality of Changxing Island river network. To make full use of the high-quality water resources, the west-to-east drainage scheme demonstrates a significantly better improvement effect on total phosphorus in the Changxing Island river network than the routine scheduling scheme, delivering greater environmental benefits. The research findings can provide technical support for promoting water circulation and water resource scheduling in Changxing Island.
[Objective] This study examines the legacy effect in aquatic nitrogen pollution control, emphasizing the role of historically accumulated nitrogen. It reviews advanced methods for quantifying lag times and legacy loads, aiming to provide a scientific basis for more precise nitrogen management. [Methods] Based on a literature review, this study analyzed nitrogen fate and transport processes, focusing on biogeochemical and hydrological legacy nitrogen. It evaluated current quantification approaches and the limitations of hydrological models. [Results] The analysis indicated that historically accumulated nitrogen could remain in watershed soils and groundwater in various forms, constituting a persistent pollution source that prevented an immediate response to management measures. Although recent research made some progress in quantifying lag times and legacy loads, current hydrological models still exhibited significant shortcomings in accurately characterizing the spatial distribution of legacy nitrogen, which limited the predictive capabilities for the lagged nitrogen response. [Conclusion] The study concludes that, to overcome the limitations of current models and effectively address the challenge posed by lag time in nitrogen pollution management, future research should focus on establishing a source-pathway coupled model for nitrogen export. This model integrates precise source identification with advanced simulation of export pathways, thereby providing a critical tool for achieving precise nitrogen management and rapid water quality improvement with minimal investment.
[Objective] During the impact of Typhoon Kong-Rey on Shanghai in the autumn of 2024, the Suzhou River reached a new historical high water level. To deeply analyze the causes of this high water level event, assess the response capacity of the existing flood control and drainage system, and explore optimized scheduling and engineering measures, this study systematically reviews the hydrological process of the high water level in the Suzhou River during Typhoon Kong-Rey and proposes practical countermeasures. It aims to provide insights and a scientific basis for optimizing flood control scheduling and urban flood control and drainage system in Shanghai, while also serving as a reference for other cities facing similar challenges. [Methods] A method combining field investigation and numerical simulation was adopted, and data on rainfall, water level, tidal level, and hydraulic facility scheduling during Typhoon Kong-Rey were collected. Considering factors such as rainfall-runoff, river network hydrodynamics, and pump-gate scheduling, a hydrodynamic model for the tidal river network was constructed to simulate the flow dynamics and water level changes in the Suzhou River and its adjacent river network. The average coefficient of determination for water level simulations reached 0.96. On this basis, a knowledge graph was utilized to identify the causes of the high water level in the Suzhou River. Three types of countermeasures were proposed: emergency discharge restriction on both banks, emergency diversion in the river network, and optimized planning for increased drainage. Different scheduling schemes were set up for simulation and comparison to quantitatively evaluate their effectiveness in reducing high water levels and their risk impacts. [Results] Simulations showed that the high water level in the Suzhou River during Typhoon Kong-Rey was primarily caused by the combined effects of concentrated rainfall in the middle and lower reaches, substantial inflow of floodwater from both banks, and the backwater effect from the high tidal level of the Huangpu River. Simulations of different countermeasures revealed the following results. (1) Emergency discharge restriction on both banks: Short-term discharge restriction in the Jiabaobei and Dianbei areas could reduce the highest water levels along the Suzhou River by 0.16-0.29 m, lowering the highest water level at Beixinjing to below 4.25 m. (2) Emergency diversion in the river network: Combining discharge restriction in Jiabaobei and Dianbei areas with emergency diversion via the Xinchapu River could maintain the highest water level along the entire Suzhou River below 4.20 m, diverting approximately 1.02 million m3 of floodwater, with minimal impact on flood control on both banks. (3) Optimized planning for increased drainage: After the implementation of the planned Suzhou River estuary pump station and Wenzaobang east pump station, the reduction in the highest water level along the Suzhou River could reach 0.40-0.64 m, while also enhancing the drainage capacity of the Jiabaobei area and significantly improving regional flood control resilience. [Conclusion] Existing engineering system for the Suzhou River has shortcomings under extreme events. Scientific scheduling and engineering optimization can effectively reduce the risk of high water levels. It is recommended to prioritize the “Jiabaobei + Dianbei discharge restriction + Xinchapu diversion” as the emergency scheduling scheme, and to accelerate the construction of the Suzhou River estuary pump station and the Wenzaobang east pump station, thereby establishing a multi-level flood control and drainage system of “restriction-diversion-expansion”. This study provides replicable and scalable scheduling experience and engineering approaches for Shanghai to cope with similar extreme typhoon events, and also offers important references for other plain cities with tidal river networks.
[Objective] Freak wave is a marine disaster characterized by extremely large wave height, strong nonlinearity, and high destructiveness. The results of wave superposition method for simulating freak waves are influenced by multiple parameters, and the sensitivity and interaction mechanisms of these factors require systematic investigation. [Methods] Based on a self-developed viscous-flow numerical wave tank, we conducted a numerical simulation on the generation of freak waves and their influencing factors. First, the reliability of the numerical model was verified against physical experimental data. Subsequently, the harmonic separation method was employed to examine the influence of wave group nonlinearity on wave surface deformation, focusing characteristics, and frequency spectrum structure. Through numerical experiments, the effects of key parameters—including spectral type, number of constituent waves, spectral bandwidth, spectral peak frequency, and water depth—were investigated. [Results] 1) During the generation of a freak wave, wave-wave nonlinear interactions caused energy to transfer from the primary frequency to both high and low frequencies, resulting in significant spectral broadening. Low-frequency free pseudo-harmonics propagated faster, leading to an actual wave height slightly larger than the theoretical value. High-frequency bound harmonics formed a tail wave, which had a minor influence on the shape of the main peak. 2) The spectral type significantly influenced the wave profile characteristics: the JONSWAP and P-M spectra, with concentrated energy, tended to generate freak waves with steep crests. The CWA spectrum produced gentle wave profiles; the CWS spectrum yielded the smallest focused amplitude. 3) The number of constituent waves affected the focusing recurrence period. An insufficient number could generate secondary focused waves. It was recommended to use 29 constituent waves to balance computational accuracy and efficiency. 4) Under finite water depth conditions, the focused amplitude reached its maximum when the spectral bandwidth was 0.7 Hz, indicating that the amplitude was co-modulated by the spectral bandwidth, water depth, and spectral peak frequency. 5) An increase in the spectral peak frequency enhanced nonlinearity, resulting in wave profile steepening. However, an excessively high frequency led to wave breaking, thereby reducing the amplitude. 6) Water depth influenced the wave profile by altering the dispersion characteristics. A greater water depth resulted in faster wave speed and a higher amplitude, whereas an excessively small water depth readily induced wave breaking. [Conclusion] The main innovations of this research include: establishing a high-precision viscous-flow numerical model capable of accurately simulating the evolution of nonlinear waves including breaking effects; employing the harmonic separation method to reveal the influence mechanism of wave group nonlinearity on wave surface structure and energy distribution; and clarifying the coupling effects of various factors under finite water depth conditions through multi-parameter sensitivity experiments. The findings of this study deepen the understanding of freak wave generation mechanisms, provide an important theoretical basis and parameter selection guidance for laboratory simulation of freak waves.
[Objective] The centrifugal and inertial forces of discharge flow in curved stilling basins lead to the impact of the flow on the concave bank and water surface rise, and result in the formation of a significant transverse water surface gradient, thereby inducing hazards threatening the structural safety, including scouring and deepening of the concave bank and sediment deposition on the convex bank. To address these challenges, this study investigates the energy dissipation and diversion characteristics of a combined energy dissipator comprising rough strips and trapezoidal piers arranged within the stilling basin. This study aims to provide a systematically validated optimized design scheme and theoretical prediction tool for engineering applications addressing complex hydraulic problems. [Methods] Taking the TGZBL Reservoir in Xinjiang as the engineering background, a 1∶60 scale physical model was constructed following the gravity similarity criterion. The model consisted of four components: a straight approach channel, a diffusion section composed of an Ogee curve and a reverse curve segment, a main curved stilling basin section incorporating an arc segment, and a discharge channel. Water depths at 35 cross-sections were measured using 0.1 mm precision point gauges, while flow velocities at two-thirds the water depth beneath the surface were recorded at left (A), middle (C), and right (E) measurement points in key cross-sections. Evaluation metrics included the energy dissipation rate η based on the Bernoulli's equation and the coefficient of variation of the transverse water surface gradient Cv to quantify the dispersion degree and assess energy dissipation and flow diversion performance. An orthogonal experimental design was employed at the core of the study, with seven influencing factors selected, including the relative width of the rough strips (b), the angle between the rough strips and the cross-section (θ), the relative height of the rough strips (h1), the height ratio of the rough strips (λ), the relative spacing of the trapezoidal piers (Δζ), the length of the trapezoidal piers (ϕ), and the longitudinal section dimensions of the trapezoidal piers (γ). Each factor was set at three levels, and 18 test conditions were arranged using the L18 (37) orthogonal array. [Results] Main effects analysis of variance (ANOVA) revealed that the most significant factors influencing the energy dissipation rate η were the trapezoidal pier spacing Δζ (P=0.005) and the relative height of the rough strips h1 (P=0.045), with the degree of influence ranked as: Δζ > h1> θ > λ > ϕ > b > γ. This was because Δζ directly altered the number and water-facing area of the trapezoidal piers, enhancing counterflow resistance and turbulent dissipation, while variations in h1 effectively redirected high-kinetic-energy flow from the concave bank toward the convex bank, maximizing the energy dissipation capacity of the latter. For the flow diversion effect Cv, the most influential factors were the relative width of the rough strips b (P=0.005), the length of the trapezoidal piers ϕ (P=0.011), the height ratio of the rough strips λ (P=0.015), and the spacing of the trapezoidal piers Δζ (P=0.023), ranked as: b > ϕ > λ > Δζ > h1 > γ > θ. Among these, b, h1, and λ collectively determined the obstructive and frictional effects of the rough strips on concave-bank flow, forcing redirection toward the convex bank and thereby achieving more uniform water depth distribution within the basin. Through post hoc multiple comparisons and comprehensive analysis of factor-level trends, the optimal parameter combination balancing both energy dissipation and flow diversion effects was determined as A2B3C3D3E3F3G1. Validation tests for this optimal configuration demonstrated significant improvements compared to the initial condition without energy dissipators: the energy dissipation rate η increased from 72.47% to 82.22% (a net gain of 9.75%). The coefficient of variation of the transverse water surface gradient Cv decreased from 0.981 9 to 0.161 2 (an 83.58% reduction). The vortex flow within the basin was mitigated, with average flow velocities at the concave and convex banks declining from 7.72 m/s and 12.42 m/s to 3.03 m/s and 3.43 m/s, respectively (reductions of 60.75% and 72.38%), and the inlet velocity of the discharge channel was substantially lowered. To translate the research findings into practical tools, a multi-factor evaluation model incorporating 11 dimensionless parameters was established based on dimensional analysis principles. Through multiple regression analysis, logarithmic function equations (adjusted R2=0.946) for predicting energy dissipation rate η and asymptotic function equations (adjusted R2=0.804) for forecasting the coefficient of variation of the transverse water surface gradient Cv were respectively developed. To ensure model reliability, six independent working conditions (including the optimal combination), which had not been involved in the fitting process, were selected for validation. The results demonstrated that the relative errors between predicted and measured values for both η and Cv remained below 10%, confirming the established semi-theoretical and semi-empirical formulas achieved excellent precision and applicability. [Conclusion] Verifications through anti-sliding and anti-overturning stability calculations confirm that the combined energy dissipator meets safety requirements under design hydraulic loads, ensuring its safety and effectiveness in practical engineering applications.
[Objective] Hydrological stations play a crucial role in monitoring hydrological regime changes. Achieving online flow calculation at hydrological stations and improving flow calculation accuracy are of great research significance for hydrological monitoring, flood and drought disaster prevention, and water resource management. [Methods] By monitoring the upstream and downstream water levels of hydrological stations as boundary conditions, a one-dimensional hydrodynamic model of the river reach near the hydrological station was constructed. The Kalman filter technique was employed to automatically calibrate the model’s roughness parameters based on measured flow data from the station. Using the upstream and downstream water levels as inputs and the automatically calibrated roughness data as outputs, a BP neural network was constructed to fit the complex relationship between water levels and model roughness. During online flow calculation, the roughness values were corrected using the real-time upstream and downstream water levels and the water level-roughness neural network relationship to improve flow calculation accuracy. By correcting the roughness based on real-time upstream and downstream water levels and using the constructed one-dimensional hydrodynamic model for simulation calculation, online flow calculation at hydrological stations was achieved. [Results] Taking Lanxi Hydrological Station as an example, the accuracy of online peak flow calculation at Lanxi Station using the proposed method was higher than that of the currently used index velocity method. For three major flood events selected, the flood flow calculation accuracy at Lanxi Station using the proposed method was higher than that of the index velocity method currently used at Lanxi Station. The reason was that the index velocity method, when establishing the relationship between index velocity and cross-sectional average velocity, used only boat-measured flow data to calibrate the relationship, which could lead to significant errors and consequently larger errors in peak flow simulation. In contrast, this study constructed a one-dimensional hydrodynamic model, used measured flow data to automatically calibrate the model roughness parameters, corrected roughness based on real-time upstream and downstream water levels to perform online flow calculation with the model, and utilized more real-time water level information than the index velocity method for model calibration and assimilation, thus achieving higher peak flow calculation accuracy. [Conclusion] This study achieves online flow calculation at hydrological stations and improves flow calculation accuracy by utilizing upstream and downstream water levels. The applicability of the method is verified using Lanxi Hydrological Station as an example, demonstrating significantly improved flow calculation accuracy compared to the index velocity method, particularly during major floods with high water levels. The method proposed in this paper is suitable for online flow calculation at hydrological stations located in upstream or midstream river reaches with relatively stable riverbed cross-sections where sediment erosion and deposition effects can be neglected. Considering the relatively low cost of water level gauges, the method demonstrates good application prospects and promotion value.
[Objective] This study aims to establish a field-based method that uses heavy dynamic cone penetration test (DCPT) energy index Pindex to quantify relative density (Dr) of sandy-cobble soil and to link Dr to deformation moduli E50 and Eur, thereby overcoming the long-standing difficulties of retrieving undisturbed samples and calibrating parameters for this material. [Methods] (1) A Φ600 mm × 600 mm calibration chamber was fabricated to enable precise reconstitution of specimens at Dr=0.40, 0.55, 0.70, and 0.85. (2) A series of 63.5 kg DCPTs were performed, and penetration resistance-depth curves were analyzed to extract Pindex. (3) Triaxial consolidated-drained and unloading-reloading tests were conducted on the same Dr specimens to obtain E50 and Eur. (4) A three-tier Bayesian-Bootstrap model “Pindex-Dr-modulus” was developed and coded into PLAXIS-Hardening-Soil model. (5) The method was validated using field monitoring data of an adjacent excavation. [Results] (1) Pindex decreased with Dr following a power law (R2≥0.93). When Dr increased from 0.40 to 0.85, Pindex dropped by 62%, outliers decreased by 47%, and repeatability error remained <3%. (2) E50=112.4 Dr^1.87 MPa and Eur=318.6 Dr ^1.64 MPa. Eur/E50 decreased exponentially from 2.8 to 2.1. (3) Ten-fold cross-validation yielded a mean absolute error of Dr =0.028. The relative errors of the predicted E50 and Eur were <8% and <7%, respectively. (4) FE simulations using the predicted moduli yielded an average relative displacement error of 6.1% compared to 18.4% (Mohr-Coulomb) and 12.7% (Modified Mohr-Coulomb), and the maximum vertical displacement deviation of the station reduced from 5.2 mm to 1.7 mm. (5) The proposed method was applicable to sandy-cobble layers in the upper and middle reaches of the Yongding River alluvial fan, western Beijing (Dr=0.35-0.90, cobble content ≤70%). [Conclusion] The study presents the first continuous field method linking DCPT impact energy, relative density, and deformation moduli for sandy-cobble soil without undisturbed sampling. The compact power-exponential model can be directly implemented in commercial software, providing in-situ parameters for deformation analysis of excavations and tunnels in such formations. The significant improvement in deformation prediction accuracy provides immediate advantages for risk control and support optimization during tunnelling or excavation near existing metro structures.
[Objective] Reverse-graded deposits are prone to geological disasters under rainfall conditions. To comprehensively understand their seepage characteristics and disaster-causing mechanism, the indoor seepage test is conducted using typical deposits as the research object. The variation patterns of permeability across different deposit layers and the migration characteristics of fine particles were studied under different fine particle contents, dry densities, and infiltration heads. [Methods] Based on the Taheman deposit landslide, constant-head seepage experiments were performed using a seepage erosion device under different fine particle contents, dry densities, and infiltration heads. The influences of external hydraulic conditions and reverse-graded deposit characteristics on the permeability of each deposit layer were analyzed, and the migration patterns of fine particles within each layer were revealed. [Results] Fine particle content, dry density, and infiltration head exerted the greatest influence on the permeability variations in the upper rock-soil layers, followed by the middle rock-soil layers, with the least influence on the lower rock-soil layers. Higher fine particle content facilitated the accumulation and deposition of fine particles in the upper and middle rock-soil layers, resulting in a decrease in the soil permeability. Under varying infiltration heads, the permeability of the upper layers exhibited either an increase or decrease, while the middle and lower layers consistently showed permeability decline. Regardless of influencing factors, the permeability of the lower layers uniformly decreased. Fine particles(<0.075 mm) had obvious migration, loss, and deposition processes in the middle and upper rock-soil layers. The lower the soil layer was, the smaller the particle size of the fine particles it could retain. [Conclusion] The middle and upper parts of reverse-graded deposits exhibit weaker erosion resistance, where fine particles are prone to migration and loss, leading to reduced permeability. The erosion resistance of the middle section is stronger than that of the upper part, while the lower section demonstrates the highest erosion resistance. Fine particles are easily subject to deposition and accumulation, which reduces permeability. Factors such as fine particle content, dry density, and infiltration head have the greatest impact on permeability variations in the upper section, followed by the middle section, with the least effect on the lower section. When the seepage direction is top-down, significant particle migration occurs between the upper and middle layers, but fewer fine particles migrate and deposit into the lower layers. The ability to retain fine particles decreases with proximity to the lower part of the deposit. Due to the coarse-upper and fine-lower structure of reverse-graded deposits, particle migration and loss characteristics vary with different seepage directions, warranting further investigation in subsequent experiments.
[Objective] Rock material typically exhibits nonlinear strength characteristics under complex loading conditions, and stress-drop can be observed during shear failure while retaining part of the residual strength. To investigate the mechanical properties of carbonaceous shale, conventional triaxial compression tests were conducted, and a new rock damage model was established based on continuum damage mechanics to describe the stress-strain curves. [Methods] The proposed model first utilized a nonlinear exponential strength criterion to describe the micro-elements of rocks, considering the material heterogeneity, and assuming the micro-element strength followed a Weibull distribution. The damage variable was derived from the accumulated failure proportion of micro-elements. Subsequently, the model employed a modified Lemaitre equivalent strain assumption to capture the stress-drop effect and residual strength, allowing for the entire stress-strain curve to be represented. Model parameters were determined using the extremum method. Finally, the model’s predictions were compared with conventional triaxial compression test results from different rock types to verify its validity. [Results] Results showed that the established rock damage model accurately described the entire stress-strain relationship of rock samples under various confining pressures. During the post-peak deformation stage, the shear strength of the rock samples dropped rapidly and eventually approached the residual strength due to the stress-drop effect, and the rock samples became fully damaged. The comparisons also suggested that the exponential strength criterion was generally suitable for various rocks; however, both the axial strain corresponding to peak strength and the residual strength varied approximately linearly with confining pressure. [Conclusion] The established exponential damage model of rock has good prospects for theoretical application.
[Objective] Due to the excellent load-bearing performance, geosynthetic reinforced soil (GRS) composites have been widely adopted in the construction of load-bearing GRS bridge abutments. Unlike conventional gravity or cantilever retaining walls, GRS abutments are required to bear significantly higher vertical loads transferred from the superstructure. Therefore, it is essential to investigate the ultimate bearing behavior of GRS composites to ensure the safety and reliability of the structures. [Methods] In this study, a series of plane strain model tests were conducted to evaluate the ultimate bearing capacity of GRS composites. Nine groups of tests were designed and conducted using geotextile as the reinforcement material, incorporating four types of backfill material gradations and three reinforcement spacings. The gradation of the backfill materials primarily varied in particle size distribution within the range of 1-8 mm, while the reinforcement spacing was set at 20 cm, 25 cm, and 33.3 cm. The test results were compared with those of unreinforced soil and analytical predictions based on the Federal Highway Administration (FHWA) design guidelines. [Results] The experimental results demonstrated that reinforcement significantly enhanced the ultimate bearing capacity of GRS composites. Under the same backfill material condition, the incorporation of reinforcement led to significant increases in ultimate bearing capacity compared with the unreinforced test. Specifically, with reinforcement spacings of 20 cm and 25 cm, the ultimate bearing capacity increased by 87.5% and 62.5%, respectively. These results clearly indicated that the reinforcement spacing played a critical role in the bearing performance of GRS composites. In addition, smaller spacings resulted in greater overall stiffness of the composite system. When the reinforcement spacing was constant and the backfill particle size ranged between 1 mm and 8 mm, the effect of gradation on the ultimate bearing capacity was relatively minor. However, differences in backfill material gradation led to noticeable variations in the overall stiffness of GRS composites. When the experimental results were compared with predictions obtained from the FHWA-recommended method for GRS composite bearing capacity, a significant discrepancy was observed. The FHWA method considerably underestimated the ultimate bearing capacity in all test cases. Therefore, it was not recommended to calculate the ultimate bearing capacity of GRS composites with finer graded backfill materials by directly applying the FHWA method. During post-test inspection, the locations of geosynthetic rupture were identified and analyzed. The observed failure surfaces within the reinforced soil mass approximately corresponded to a Rankine failure plane. The results indicated that the obvious composite behaviors were demonstrated in the GRS composites. [Conclusion] This experimental study provides a systematic analysis of the ultimate bearing capacity of GRS composites under plane strain conditions, emphasizing the roles of reinforcement spacing and backfill material gradation. The findings confirm that geosynthetic reinforcement can significantly enhance both the strength and stiffness of the soil composites, with closer reinforcement spacing resulting in better performance. The study reveals that the current design guidelines recommended by FHWA significantly underestimate the actual ultimate bearing capacity, particularly when the backfill material gradation differs from the recommended values. These findings offer valuable reference for future engineering design and construction, promoting more efficient and reliable use of fine-grained or narrowly graded soil in reinforced soil structures.
[Objective] This study aims to systematically investigate the control measures for geyser intensity in baffle-drop shafts during the release of high-pressure trapped air pockets. By analyzing the effects of key parameters—including the connection mode of the communication pipe, the area of the connecting region between dry and wet zones, the position and open area of the throttling orifice plate, and the installation distance of the vent pipe—on the geyser height and the impact load on baffles, a set of comprehensive optimization measures balancing geyser control effectiveness and structural safety is proposed. [Methods] FLUENT software was used to establish a three-dimensional numerical model of geyser in a baffle-drop shaft based on the Realizable k-ε turbulence model and the VOF two-phase flow model. The Dongfeng Road baffle-drop shaft of the Donghao Chong deep tunnel project in Guangzhou was selected as the research object. The effects of different communication pipe connection modes (dry zone/wet zone), areas of the connecting region between dry and wet zones, throttling orifice plate parameters (height and open area), and vent pipe installation distance on the jet height of geysers and the impact load on baffles were systematically simulated. A total of 88 working conditions were simulated, and model reliability and computational accuracy were ensured through grid independence verification and comparison with experimental data. The response patterns of geyser intensity and baffle impact load to each parameter were analyzed in detail. [Results] Although connecting the communication pipe to the wet zone had a limited effect on the geyser height, it significantly reduced the impact load on the baffles—particularly on the bottom baffle, where the peak load was reduced by up to 66%. The area of the connecting region between the dry and wet zones showed a nonlinear relationship with the baffle load; as the area decreased, the impact load on the bottom baffle increased markedly. The optimal control effect was achieved when the dimensionless area S*=0.318. When the throttling orifice plate was positioned at the mid-height of the dry zone (1/2H) with an open area of ϕ*=0.058, the maximum geyser height decreased by approximately 70%, while the impact load on the baffles dropped by more than 30%. The best control effect was achieved when the vent pipe was positioned at the end of the communication pipe farthest from the shaft (δ*=4D), and no water-air mixture overflow occurred. [Conclusion] Considering the combined influence of these factors on the geyser intensity in the shaft, a joint control measure—“wet-zone connection + mid-position throttling orifice plate + remote vent pipe + optimized connecting area layout”—was proposed. Under typical working conditions, this combined approach reduced the geyser height by up to 80% and the average impact load on the baffles by more than 50%, effectively controlling the geyser intensity while ensuring the structural safety of the shaft. It provides reliable theoretical support and practical guidance for the safe design and operation of baffle-drop shafts in deep tunnel drainage systems and offers a replicable and scalable technical approach for future geyser risk prevention and control in deep tunnel systems.
[Objective] The 3D laser scanning technology is characterized by fast scanning speed, high scanning accuracy, non-contact operation, and minimal influence from the scanning environment, which makes it widely applicable in deep engineering fields. However, high in-situ stress and complex geological structures result in complex tunnel surface morphology and a non-linear actual axis, making the filtering and classification of point clouds for deeply buried tunnels more difficult than those for shallow-buried projects. This study aims to address the recognition and classification of point cloud profile for deeply buried irregular tunnels. [Methods] Based on the spatial morphology of the contour of deeply buried irregular tunnel, we established a two-level filtering method for the point cloud of deeply buried irregular tunnels, and developed a tunnel point cloud classification method based on density-based clustering and spatial position classification. [Results] To verify the effectiveness of the proposed methods, the point cloud data of a 30 m-long deep tunnel excavated by the drilling and blasting method were used as the research object. First, two 1 m-long tunnel segment point clouds were selected for analysis. The filtering effects of the tunnel segment point clouds were analyzed under different segment thicknesses L=0.2, 0.4, 0.6 m and distance thresholds dcritical=0.02, 0.04, 0.06, 0.08,0.1 m. Through comprehensive comparison, the optimal parameters were determined as L=0.2 m and dcritical=0.04 m, which were successfully applied in the filtering of the tunnel point cloud. On this basis, according to the spatial distribution characteristics of the tunnel segment point clouds, the DBSCAN algorithm parameters were set to ε=0.1 m and MinPts=50, which enabled the classification of non-profile point clouds of tunnel segments. [Conclusion] This study focuses on the filtering problem of point clouds in deeply buried tunnels. Based on the spatial geometric features of tunnel point clouds, a two-level filtering and classification method for point clouds of deeply buried tunnels with complex morphology is proposed. Case analysis shows that the proposed method realizes effective filtering and classification of tunnel point clouds with complex morphology, solves the recognition problem of contour point clouds in deeply buried tunnels, and provides reliable technical support for applications of 3D laser scanning point clouds in potential risk area identification, lining thickness detection, spatiotemporal deformation monitoring, and health condition assessment of deeply buried tunnels.
[Objective] This study aims to optimize the coupling conditions for CO2 reinforcement of recycled coarse aggregate (RCA) by investigating the interactions of three key factors—CO2 concentration, carbonation temperature, and relative humidity—using response surface methodology (RSM). The innovation lies in using a systematic RSM-based approach to model and optimize the carbonation process, overcoming the limitations of traditional methods by capturing complex inter-factor interactions. This provides a more efficient and reliable framework for enhancing RCA performance in sustainable construction applications. [Methods] A Box-Behnken design using Design-Expert software was applied, encompassing 17 sets of carbonation tests to evaluate the effects of CO2 concentration (20%-60%), carbonation temperature (20-60℃), and relative humidity (35%-65%). RCA was derived from waste concrete blocks in Baotou, China, and characterized in accordance with GB/T 25177-2010, with particle sizes between 4.75 and 31.5 mm. The measured responses included crush value (indicator of mechanical strength), water absorption (indicator of porosity), and apparent density (indicator of compactness). Carbonation experiments were performed in a controlled environment, and the obtained data were utilized to develop quadratic regression models using RSM. Analysis of variance (ANOVA) was conducted to assess the significance, reliability, and interactions of the models, using evaluation criteria including F-statistic, p-value, coefficient of determination (R2), adjusted R2, predicted R2, coefficient of variation (CV), and signal-to-noise ratio (Adeq Precision). Optimization was performed using the numerical module of Design-Expert to determine the optimal carbonation conditions, which were validated experimentally to confirm model accuracy. [Results] The interaction between carbonation temperature and relative humidity had the strongest effect (p>0.05 for BC interaction), followed by the CO2 concentration-temperature (AB) and CO2 concentration-relative humidity (AC) interactions. The CO2 concentration-temperature (AB) interaction was the most significant, resulting in a parabolic response. Water absorption initially decreased with increasing CO2 concentration and temperature, but increased under extreme conditions due to reduced CO2 diffusion and calcium ion dissolution. The CO2 concentration-relative humidity (AC) interaction was the most significant, making apparent density peak under moderate conditions (e.g., 42% CO2 concentration and 44 ℃) and decline at extremes due to moisture-induced calcium loss or CO2 saturation. The optimization process determined the optimal carbonation conditions as 38% CO2 concentration, 41 ℃ carbonation temperature, and 49% relative humidity. Under these conditions, the predicted values were 14.3% for crush value, 3.80% for water absorption, and 2 700 kg/m3 for apparent density. Experimental validation produced measured values of 14.6% (crush value), 3.85% (water absorption), and 2 702 kg/m3 (apparent density), with relative errors of 2.1%, 1.3%, and 0.1%, respectively. All relative errors were below 5%, confirming model accuracy. Compared with untreated RCA, the optimized carbonation treatment reduced crush value by 18.0%, decreased water absorption by 20.5%, and increased apparent density by 0.9%, demonstrating practical effectiveness. Response surface diagrams and contour plots illustrated these interactions. For example, the temperature-relative humidity interaction for crush value showed a steep elliptical contour, while the CO2 concentration-relative humidity interaction for apparent density presented a flat parabolic surface. These results highlighted the innovation of applying RSM to decipher complex multi-factor couplings, which previous studies did not fully address. [Conclusion] This study successfully develops and validates RSM-based regression models for optimizing the CO2 reinforcement of RCA, with high reliability and precision confirmed by statistical indicators and experimental validation. The optimal conditions effectively improve RCA properties and provide a sustainable solution for waste concrete recycling and carbon emission reduction. The developed model offers a reliable reference for industrial applications, facilitating the adoption of CO2 -modified RCA in concrete production. Future research can apply this approach to other aggregate types or larger-scale scenarios, further advancing a circular economy in the construction industry.
[Objective] This study aims to optimize the two-dimensional adaptive analysis strategy of the independent cover-based manifold method, focusing on addressing its deficiencies in error control and mesh distribution, thereby significantly enhancing computational efficiency and engineering practicality. [Methods] Based on the arbitrarily shaped and connected cover meshes of the independent cover-based manifold method, a “split-one-into-two” mesh splitting algorithm was employed for arbitrary refinement, and the degree of continuity of physical field derivatives was adopted as the error control indicator, forming an adaptive analysis strategy. An optimization scheme was proposed. 1) Adopting an absolute error indicator to replace the relative error indicator: the original relative error indicator tended to cause over-refinement in regions of minor stress and was overly sensitive in concave corner singularity regions. Using the absolute error indicator not only simplified the error judgment logic but also permitted larger error thresholds to be set near singular points such as concave corners, thereby effectively avoiding over-refinement. 2) Introducing a local mesh pre-partitioning and short strip elimination strategy: to address the issue of excessively high mesh density and irregular distribution in concave corner regions, a local pre-partitioning strategy was proposed, which pre-set the initial mesh in these regions by inwardly offsetting and reversely extending the edges of the concave corner. Simultaneously, an adjacent point merging algorithm was introduced during the mesh splitting process, which avoided the generation of extremely short connection strips and improved the conditioning of the system equations. [Results] Verification through two typical hydraulic structure examples, the square-hole and the gravity-dam model, demonstrated that the optimized scheme achieved a breakthrough improvement in computational efficiency. For the square-hole example, the original adaptive strategy generated 310 covers, corresponding to 6 520 degrees of freedom (DOFs). Under the same accuracy objective, the optimized scheme required only 59 covers and 933 DOFs. This represented a reduction of approximately 81% in the number of meshes and approximately 86% in DOFs. For the gravity-dam example, the original strategy generated 228 covers and 4 810 DOFs, whereas the optimized scheme required only 106 covers and 2 354 DOFs, achieving significant results of over 53% reduction in the number of meshes and 51% reduction in DOFs. The most notable achievement of the optimized scheme was in the effective suppression of mesh over-refinement near concave corner singularity regions. The calculation results demonstrated that the new strategy could generate more reasonable meshes, while ensuring computational accuracy, it substantially reduced the computational scale, and greatly enhanced the computational efficiency. [Conclusion] The proposed optimization strategy significantly enhances the efficiency of adaptive analysis while maintaining high accuracy. Through absolute error control and local mesh pre-partitioning, it effectively solves the problems of mesh over-refinement and unreasonable distribution near concave corner singularities, laying a foundation for subsequent three-dimensional adaptive analysis and engineering applications. Future research includes: criteria for selecting error thresholds and the highest order of cover series; further automating the local mesh pre-partitioning process to enable it to handle more complex geometries, ultimately achieving the goal of efficient and fully automatic simulation analysis for hydraulic structures.
[Objective] Storm sewers face challenges from both hydraulic pressures and pollutant accumulation. Exogenous inflows carry substantial pollutant loads into storm sewers, and during rainfall events, these pollutants are rapidly mobilized and discharged at outfalls, sharply degrading receiving water quality and presenting a critical environmental management challenge. This study systematically investigates the sources, dynamics, and impacts of storm sewer pollution in a representative urban drainage area with known exogenous pollution inputs. Specifically, it aims to identify dominant pollution sources, analyze the spatiotemporal evolution of pollutant accumulation during dry and rainy seasons, and evaluate the effectiveness of outlet-based control strategies through scenario simulation. The novelty of this study lies in its integration of long-term, high-frequency monitoring with multivariate statistical analysis and rainfall scenario modeling, providing mechanistic insights into storm sewer pollution processes and their mitigation. [Methods] The study was conducted in a storm sewer drainage catchment located in a densely urbanized district, characterized by frequent illicit sewer connections and adjacent industrial facilities. Comprehensive monitoring was carried out in a full hydrological year, covering both dry and rainy seasons to capture seasonal variations. The measured parameters included flow rate, COD, NH3-N, total phosphorus (TP), and conductivity, while event-based sampling during rainfall events captured the dynamics of pollutant wash-off and resuspension. To separate the contributions of different sources, principal component analysis (PCA) was applied to the monitoring dataset to identify dominant pollution signatures. Stepwise statistical methods were employed to evaluate temporal accumulation patterns. Hydrological and water quality models were developed to reproduce rainfall-runoff processes under typical storms, incorporating outlet-based control measures like initial flow detention and sedimentation. By integrating real-time data collection, multivariate source apportionment, and scenario simulations, this approach established a robust framework to analyze the interactions between exogenous pollution sources and in-pipe pollutant dynamics. [Results] The results revealed interactions between continuous exogenous inflows, internal accumulation, and rainfall-induced pollutant mobilization. PCA distinguished three dominant sources: illicit sanitary sewage connections that caused continuous baseflow contamination during dry seasons; rainfall-driven surface wash-off that produced sharp concentration spikes at the onset of storms; and industrial discharges, typically episodic with specific COD fractions and heavy metal signatures. During dry seasons, a stepwise increase in pollutant concentrations was observed, as COD and NH3-N accumulated in the sewer network due to ongoing illicit inflows and limited hydraulic flushing, creating latent pollution risks, abruptly released during subsequent rainfall events. High-frequency monitoring during storms confirmed a pronounced first-flush effect, with pollutant concentrations peaking within the first 20-30 minutes of rainfall due to both surface wash-off and the resuspension of in-pipe deposits. Scenario-based modeling demonstrated that targeted outlet control measures could significantly mitigate these impacts. The simulations showed that detaining approximately the initial 30% of stormwater outflow reduced COD and NH3-N loads by 40%-60%, highlighting the effectiveness of simple, targeted interventions. Furthermore, detention measures delayed pollutant peaks, reducing acute stress on receiving waters and improving the resilience of the urban drainage system to pollution shocks. These results underscored the importance of addressing both continuous exogenous inflows and event-based flushing dynamics for effective pollution control. [Conclusion] This study provides new evidence on the mechanisms and management of storm sewer pollution, demonstrating that internal sewer pollution results from the combination of continuous accumulation during dry seasons and sudden mobilization during rainfall events. Unlike previous studies focusing on rainfall-runoff wash-off as the primary driver, this study highlights the critical role of illicit sanitary connections and industrial discharges in maintaining persistent internal pollutant loads. By combining long-term, high-frequency monitoring with PCA-based source apportionment and rainfall scenario simulations, the study presents an innovative framework applicable to other urban areas for revealing hidden pollution dynamics. The practical implications are significant. Localized outlet control measures, particularly initial stormwater detention, provide a cost-effective and technically feasible way to reduce pollutant loads in receiving waters. On a broader scale, the findings emphasize the necessity of integrating internal storm sewer pollution management into urban water quality strategies. This study advances the understanding of pollutant accumulation and mobilization in storm sewers while providing guidance for sustainable urban drainage planning.
[Objective] Accurate prediction of influent ammonia nitrogen concentration is a key support for ensuring the stability of the biological treatment process in wastewater treatment plants and achieving low-carbon and efficient nitrogen removal. However, target wastewater treatment plants often face a shortage of effective samples due to insufficient monitoring equipment and limited data collection, which leads to poor prediction performance as traditional models are prone to underfitting or overfitting. This study aims to construct a 1DCNN-LSTM deep hybrid model based on transfer learning to overcome the bottleneck of influent ammonia nitrogen prediction under few-shot scenarios and achieve accurate prediction. [Methods] Two wastewater treatment plants with the same process in southeastern China were selected as the source domain and target domain. The source domain comprised hourly monitoring data from February to November 2024, including influent flow rate (Q), pH value, chemical oxygen demand (COD), etc., while the target domain consisted of scarce data from November 10 to 30, 2024. A 1DCNN-LSTM model was constructed, using historical data combined with multi-scale autocorrelation features and first-order difference features of ammonia nitrogen as the model input. Bayesian optimization was used to determine the model hyperparameters. Additionally, the model was first trained on the source domain. For the target domain, transfer learning was applied using a two-stage transfer strategy. First, the convolutional layers and LSTM layers of the source domain pre-trained model were frozen, and only the fully connected layers of the target domain were trained. Then, the entire model was fine-tuned with a small learning rate. Finally, performance was evaluated using indicators such as RMSE, MAPE, and R2. [Results] First, the data distributions of the source and target domains exhibited certain similarities while also showing certain differences, which conformed to the application scenario of transfer learning. The source domain model showed excellent performance, with RMSE=1.65, MAPE=4.60%, and R2=0.91 on the test set, and it could accurately capture the short-term fluctuations and long-term trends of ammonia nitrogen concentration. In the target domain, the performance of the transfer learning model was significantly better than the directly trained model. RMSE decreased from 1.650 to 1.515, a reduction of 8.18%. MAPE decreased from 5.62% to 5.21%, a reduction of 7.23%. R2 increased from 0.635 to 0.692, an increase of 9.02%. The prediction curve of the transfer model was smoother and aligned more closely with the measured values, demonstrating stronger adaptability and stability, particularly during sudden changes in ammonia nitrogen concentration. [Conclusion] The core innovations of this study are reflected in two aspects. First, this study proposes a 1DCNN-LSTM hybrid architecture that integrates the advantages of local feature extraction and long-term dependency modeling, overcoming the limitations of single models in capturing the complex dynamic changes in ammonia nitrogen. Second, it designs a two-stage transfer strategy that not only preserves the general knowledge learned from the source domain but also adapts to the differences of the target domain through fine-tuning, effectively addressing the issues of small samples and domain shift and avoiding the accuracy decline caused by directly applying the source domain model. The results confirm that the 1DCNN-LSTM model can reliably capture the variation patterns of ammonia nitrogen, and transfer learning can significantly enhance the prediction accuracy and generalization ability under few-shot scenarios. This provides a reliable technical pathway for wastewater treatment plants to precisely regulate process parameters and optimize chemical dosing and offers a new perspective for addressing the issue of scarce water quality monitoring data, which is of great significance for promoting the intelligent and precise wastewater treatment.
[Objective] This study investigates the discharge characteristics of antibiotics in wastewater treatment plant (WWTP) effluents and evaluates the removal efficiency of constructed wetlands (CWs) in polder areas of the middle and lower Yangtze River Basin. By systematically examining the occurrence, seasonal variation, and removal pathways of antibiotics, this study aims to provide a scientific basis for controlling antibiotic pollution in treated wastewater. Furthermore, it seeks to provide theoretical and practical insights into the mechanisms by which CWs eliminate emerging contaminants, thereby supporting the optimization of wetland design and operational strategies for enhanced pollutant removal. [Methods] A comprehensive literature review was conducted to frame the current state of knowledge regarding antibiotic pollution and wetland treatment efficacy. Liquid chromatography-mass spectrometry (LC-MS/MS) was employed to quantitatively analyze 31 antibiotics across five classes detected in the WWTP effluents. The spatiotemporal distribution patterns of these antibiotics were examined, along with the key influencing factors and their underlying mechanisms. In addition, the treatment performance of different CW units—including enhanced treatment ponds, horizontal subsurface flow wetlands, surface flow wetlands, and advanced purification ponds—was systematically evaluated. To assess the potential ecological impact, the risk quotient (RQ) method was applied to the top 10 antibiotics based on concentration in both the influent and effluent of the wetland system. This approach enabled a detailed evaluation of the ecological risks posed by antibiotic residues before and after wetland treatment. [Results] The analysis revealed a diverse profile of antibiotics in the WWTP effluents, with a total of 31 compounds across five classes detected. Macrolides, fluoroquinolones, and sulfonamides were the predominant classes. Antibiotic concentrations exhibited significant seasonal variations, influenced by the intensity of anthropogenic activities within the service area. The highest concentrations were observed in winter, followed by spring, with relatively lower levels in summer and autumn. CWs demonstrated a notable capacity for antibiotic removal, with an overall efficiency ranging from 47.29% to 65.90%. The removal rates across the treatment units were 34.20% for the enhanced treatment pond, 21.63% for the horizontal subsurface flow wetland, and 38.49% for the surface flow wetland combined with the advanced purification pond. The variations in removal efficiency across units were closely associated with the physicochemical properties of the antibiotics, such as hydrophobicity, biodegradability, and sorption potential. Ecological risk assessment based on RQ indicated that CWs effectively reduced the ecological risks of antibiotics. The RQ values for most individual antibiotics were lower in the effluent than in the influent, confirming the role of wetlands in mitigating the environmental impact of antibiotic discharges. [Conclusion] This study provides a systematic analysis of the occurrence and removal of antibiotics in WWTP effluents and CWs within the polder areas of the middle and lower Yangtze River. The findings are derived from sampling at specific locations and time points, which may not fully represent the dynamic and complex behavior of antibiotics across seasonal and hydrological cycles. Future research can expand the spatial scope to include a wider range of typical WWTPs and wetland systems across the region, thereby enhancing the generalizability and mechanistic insight of the findings. Moreover, future studies should investigate the interactions among antibiotics and other coexisting pollutants—such as heavy metals, microplastics, and antibiotic resistance genes—as well as their combined effects and removal mechanisms in CWs. These efforts will contribute to the development of more effective and reliable nature-based solutions for controlling emerging contaminants in vulnerable water environments.
[Objective] The wastewater treatment process exhibits highly non-linear, time-varying, and multivariable coupling characteristics, making it difficult for traditional prediction methods to effectively capture complex spatiotemporal dependencies. Unidirectional LSTM utilizes only historical information, struggling to fully exploit bidirectional temporal features. This study aims to construct a deep learning model combining a bidirectional long short-term memory network with an attention mechanism to achieve high-precision prediction of effluent COD in wastewater treatment plants. [Methods] This study proposed a deep learning architecture integrating BiLSTM and a multi-layer attention mechanism. The model adopted a hierarchical design. First, sine-cosine positional encoding was used to embed time step position information. A feature attention mechanism was designed to achieve adaptive weight learning for different water quality parameters using a fully connected network and the softmax function. Then, a single-layer bidirectional LSTM structure was employed to simultaneously capture forward and backward temporal dependencies. A multi-head attention mechanism was introduced to capture complex interaction patterns between time steps. Subsequently, a time-step importance weighting mechanism was designed, using a quadratic growth curve to assign higher weights to recent time steps. An attention-gated fusion strategy was used to dynamically combine the LSTM output and the attention output. Finally, the final prediction was achieved through global average pooling and a fully connected network. The model training employed the Adam optimizer, Dropout regularization, L2 regularization, and an early stopping strategy. The prediction performance was compared with baseline models such as unidirectional LSTM, BiLSTM, and 1D-CNN. [Results] Experimental verification showed that the BiLSTM-attention mechanism model significantly outperformed other models in effluent COD prediction. Compared to the BiLSTM model, the root mean square error decreased from 1.17 mg/L to 1.01 mg/L, a reduction of 13.5%. The mean absolute error decreased from 0.92 mg/L to 0.80 mg/L, a reduction of 13.8%. The mean absolute percentage error decreased from 9.79% to 8.29%, a reduction of 15.3%. The validation set loss converged well during the training process. The visualization analysis of attention weights revealed the model’s decision-making mechanism as follows. Feature attention identified dissolved oxygen in the process section and sludge concentration as key influencing parameters. Temporal attention showed that the model assigned higher weights to recent time steps, conforming to the physical laws of time-series prediction, and the different heads of the multi-head attention captured different temporal dependency patterns, achieving complementary feature extraction. [Conclusion] This study successfully constructs an effluent COD prediction model for wastewater treatment plants based on BiLSTM and a multi-layer attention mechanism. The innovations are reflected in proposing a hierarchical deep learning architecture that integrates positional encoding, feature attention, multi-head attention, and gated fusion; utilizing a bidirectional LSTM structure to simultaneously leverage forward and backward temporal information, which reduces the error by over 10% compared to unidirectional models; and designing time-step importance weighting and gated fusion mechanisms to achieve refined modeling of temporal information.
[Objective] Aeration represents the most energy-intensive process in wastewater treatment plants (WWTPs), accounting for over 50% of total energy consumption. Traditional operation strategies often rely on excessive aeration to ensure effluent compliance, causing substantial energy waste and increased carbon emissions. To address this challenge, we propose a data-driven probabilistic optimization framework integrating Gaussian process regression (GPR) with an improved black widow optimization (IBWO) algorithm to minimize aeration energy consumption while ensuring effluent quality compliance under dynamic and uncertain influent conditions. The novelty lies in combining a probabilistic prediction model with an enhanced evolutionary algorithm for adaptive and energy-efficient aeration control at the plant scale.[Methods] A GPR model was developed to describe the nonlinear relationships between influent characteristics, aeration flow rate, and effluent water quality indicators, and furthermore, an IBWO algorithm was constructed by enhancing the standard black widow optimization framework. The GPR-IBWO framework was executed in a closed-loop configuration, where the GPR model continuously updated effluent predictions from real-time process data, and IBWO dynamically optimized aeration flow rates to minimize energy consumption while maintaining effluent quality within probabilistic bounds. Benchmark tests on six CEC2017 functions demonstrated that IBWO outperformed GA, PSO, ABC, and standard BWO in convergence speed and solution accuracy. The GPR-IBWO strategy was tested in a full-scale municipal WWTP in Jiujiang City, China. Operational data were collected over 30 days during wet and dry seasons to capture hydraulic and load fluctuations. The performance of four strategies was compared: the proposed GPR-IBWO optimization, conventional plant control mode, baseline GPR-BWO strategy method, and benchmark LSTM-PSO framework. [Results] Under both hydrological conditions, GPR-IBWO consistently maintained effluent concentrations of N
[Objective] Traditional hydrological models are mostly designed for specific scenarios, representing simplifications of complex physical processes or fitting of historical data. They require extensive modeling and driving data, and have high technical threshold for use, which largely limits their widespread application. Large language models, with their powerful understanding and generative capabilities, demonstrate superiority in the framework design and interaction processes of complex systems. Integrating large language models with hydrological model systems can provide an intelligent engine for the data collection, model construction and calibration, and simulation result analysis required by hydrological models. This enhances the adaptability and extensibility of hydrological models and lowers the model usage threshold. Therefore, this study explores the synergistic application paths for large language models and hydrological models and provides a prospect for future research priorities in their synergy. It aims to provide support and technical reference for the synergistic development of the two models, reduce the usage threshold of hydrological models, and accelerate the intelligent transformation of hydrological science. [Methods] Starting from hydrological models, based on the collection and organization of current hydrological model information and the classified elaboration of their characteristics, this study identifies problems such as poor portability, sensitivity to data quality, and weak predictive capability for extreme events. Additionally, it clarifies three requirements for their intelligent transformation: integrating knowledge-assisted decision-making, optimizing interaction methods to lower the usage threshold, and enhancing multi-model synergy to address complex problems. Combining the current research and application status of large language models, this study reviews the technical trends of mainstream large language models and analyzes the challenges in their practical application, including a lack of domain-specific knowledge, weak business reasoning capability, and difficulties in collaborating with professional tools. Finally, the study points out that hydrological models and large language models have complementary advantages, making their coupling inevitable. An application example is provided, focusing on the composite hydrological process of lake inflow from natural and urban underlying surfaces. [Results] This study proposes synergistic approaches for unidirectional and bidirectional coupling between large language models and hydrological models. This enables the effective utilization of large language models across various application scenarios of hydrological models, provides support and technical reference for their synergistic development, and contributes to lowering the usage threshold of hydrological models and accelerating the intelligent transformation of hydrological science. Combining the synergistic approaches for unidirectional coupling and bidirectional coupling, and based on the composite hydrological process of lake inflow from natural and urban underlying surfaces, a conceptual framework for coupling large language models and hydrological models that is universal and innovative is constructed. Additionally, specific implementation procedures are provided, offering a reusable methodological reference for studying complex hydrological processes. [Conclusion] The coupling technology of large language models with hydrological models, through continuous exploration and research, has preliminarily demonstrated the application potential of integrating large language models with hydrological simulation and prediction. Large language models, leveraging their advantages in semantic parsing, dynamic parameter calibration, and multi-model synergistic scheduling, have expanded the functional boundaries of hydrological models in aspects such as real-time interactive response, cross-scale coupled simulation, and emergency decision-making support. This integration not only lowers the application threshold of professional hydrological models through intelligent interfaces, but also enhances the efficiency of simulating complex hydrological processes and the accuracy of emergency decision-making through dynamic parameter optimization and synergistic computing mechanisms. The coupling of these two models demonstrates rapid adaptability to novel environmental conditions and the characteristics of autonomous reasoning and optimization, paving an innovative path for establishing a “perception-simulation-decision” full-chain integrated digital twin watershed system.
