Search Results (3,597)

Water inrush and mud burst disasters pose severe challenges to the safe and efficient construction of underground engineering. Water inflow prediction is closely related to drainage design, disaster prevention and control, and the safety of the surrounding ecological environment. Thus, assessing the water inflow accurately is of importance. This study proposes a Bayesian Optimization-Long Short-Term Memory (BOA-LSTM) recurrent neural network for predicting tunnel water inflow. The model is based on four input parameters, namely tunnel depth (H), groundwater level (h), rock quality designation (RQD), and water-richness (W), with water inflow (WI) as the single-output variable. The model first processes and analyzes the data, quantitatively characterizing the correlations between input parameters. The tunnel water inflow is predicted using the long short-term memory (LSTM) recurrent neural network, and the Bayesian optimization algorithm (BOA) is employed to select the hyperparameters of the LSTM, primarily including the number of hidden layer units, initial learning rate, and L2 regularization coefficient. The modeling process incorporates a five-fold cross-validation strategy for dataset partitioning, which effectively mitigates overfitting risks and enhances the model’s generalization capability. After a comprehensive comparison among a series of machine learning models, including a long short-term memory recurrent neural network (LSTM), random forest (RF), back propagation neural network (BP), extreme learning machine (ELM), radial basis function neural network (RBFNN), least squares support vector machine (LIBSVM), and convolutional neural network (CNN), BOA-LSTM performed excellently. The proposed BOA-LSTM model substantially surpasses the standard LSTM and other comparative models in tunnel water inflow prediction, demonstrating superior performance in both accuracy and generalization. Hence, it provides a reference basis for tunnel engineering water inflow prediction. Full article

The present research moves the focus from merely describing epikarst flow to quantifying its natural filtration performance and contaminant retention mechanisms through integrating in situ tracer experiments with controlled laboratory modelling—an approach seldom applied in previous studies. Two field experiments at Peč Cave demonstrated that the epikarst exhibits rapid hydraulic connectivity—evidenced by fast tracer breakthrough with virtual flow speeds between 0.0041 and 0.006 m/s—yet simultaneously provides strong attenuation, as shown by the low tracer recovery and near-complete removal of microbial contaminants as well as nitrogen compounds through retention, degradation, and dilution under natural infiltration conditions, including rainfall and snowmelt. Complementary laboratory simulations further confirmed this duality, with nitrate concentrations reduced by 30–50%. Field data and lab results consistently indicated that the epikarst does not merely transmit water but actively adsorbs and transforms pollutants. Overall, the epikarst on Suva Planina functions as an effective natural filtration layer that substantially improves groundwater quality before it reaches major karst springs, acting as a protective yet vulnerable “skin” of the aquifer. These findings highlight the epikarst’s critical role in Suva planina Mt. karst aquifer protection and results support consideration of epikarst in groundwater management strategies, particularly in regions where springs are used for public water supply. Full article

Understanding water and heat transport through soils is vital for managing soil and groundwater resources, agricultural irrigation, and ecosystem protection. This paper aims to explore the potential application of deep learning methods in simulating water and heat transport processes within soils. It also examines the interactions between soil hydrological processes and environmental factors, including meteorological conditions and groundwater levels. To achieve these, we develop a hybrid model Informer–LSTM by combining two powerful architectures: Informer, a Transformer-based model essentially designed for long-sequence time-series forecasting, and Long Short-Term Memory (LSTM), a neural network that is great at learning short-term patterns in sequential data. The model is applied to field measurements from Henan Township in Ordos, Inner Mongolia, China, for training and testing, to simulate three key variables: soil water content, temperature, and heat flux at different depths in two soil columns with different groundwater levels. Our results confirm that Informer–LSTM is highly effective at simulating the soil water and heat transport. Simultaneously, we evaluate its performance by incorporating various combinations of input data including meteorological data, soil hydrothermal dynamics, and groundwater level. This reveals the relationship between soil hydrothermal processes and meteorological data, as well as coupled processes of soil water and heat transport. Moreover, employing SHapley Additive exPlanations (SHAP) analysis, we identify the most influential factors for predicting heat flux in shallow soils. This research demonstrates that deep learning models are a viable and valuable tool for simulating soil hydrothermal processes in arid and semi-arid regions. Full article

Proactive management of groundwater resources is hindered by the static nature of conventional vulnerability assessments, which provide only a single temporal snapshot and lack predictive capability. To address this limitation, we developed a coupled dynamic–spatial modeling framework to forecast the spatio-temporal evolution of groundwater vulnerability. The framework integrates a βSARMA time-series model for precipitation forecasting with an enhanced M-DRASTIC-LAaRd model, which incorporates Land use, Anthropogenic activity, and River network density, weighted via the Analytical Hierarchy Process (AHP) to better capture hydrogeological complexity. The βSARMA model consistently outperformed conventional SARIMA models across the five subregions of Beijing, achieving the lowest RMSE values (0.0832–0.1617) and MAE values (0.0922–0.1372), with an average RMSE reduction of 15.3% relative to the best SARIMA baseline. These results ensure highly reliable dynamic precipitation inputs for the time-varying Net Recharge (R) parameter. Model validation against historical observations yielded a coefficient of determination (R²) of 0.87, confirming the framework’s robustness and predictive accuracy. Applied to the Beijing metropolitan area (1980–2027), the model projects a marked spatial restructuring of groundwater vulnerability: high-vulnerability zones are expected to expand from 38.65% to 46.18%, while low-vulnerability areas will decline from 42.53% to 34.63%. Emerging “hotspots” are concentrated in the southern urban plains, where urbanization and reduced recharge converge. Overall, 27.9% of the region is predicted to experience intensified vulnerability, whereas only 11.5% will show improvement. This study advances groundwater vulnerability assessment from static mapping toward dynamic forecasting, providing a quantitatively validated and spatially explicit framework that supports more informed groundwater management under future environmental change. Full article

Groundwater management in volcanic islands represents a complex challenge due to the scarcity of surface resources, the strong heterogeneity of volcanic terrains, and the constant threat of marine intrusion. In Tenerife (Canary Islands, Spain), current regulations establish that only saline or brackish waters are permitted for exploitation, to be subsequently desalinated through reverse osmosis for urban and touristic supply. In this context, it is essential to develop geophysical methodologies capable of accurately characterizing subsurface salinity and optimizing the location of new boreholes. The present study applies Electrical Resistivity Tomography (ERT) profiles in the Los Cristianos area (Arona, Tenerife), later integrated into a three-dimensional model using Oasis Montaj software Version 2025.1. The results allow for the differentiation of four geoelectrical domains. The 3D modeling enabled a detailed characterization of the conductive domain, delineating the geometry of the marine intrusion. The findings confirm that the combination of ERT and 3D modeling constitutes an effective, replicable, and economically efficient methodology for precisely locating saline horizons and selecting the most suitable drilling sites, thereby providing an objective basis for the sustainable management of water resources in volcanic islands. Full article

A global transition in water management is currently underway, marked by the declining reliability of conventional sources and the accelerated adoption of non-conventional alternatives. This shift is driven by escalating pressures from climate change, population growth, and freshwater overexploitation. While the literature on management of water sources (WSs) is extensive, empirical clarity on Hybrid Water Systems Management (HWSM)—the integration of conventional and non-conventional WSs within a single system—remains limited. The present study addresses this gap through a systematic literature review using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) approach, which ensures methodological transparency and applicability. From over 9000+ peer-refereed articles retrieved from three major scientific databases (ScienceDirect, Scopus, and Web of Science Core Collections), published between 1999 and 2024, 44 studies were identified as the most relevant and consequently analyzed. The literature review refines the classification of WSs, distinguishing conventional sources, such as groundwater and surface water, from non-conventional alternatives, such as desalinated water, treated wastewater, gray water, and rainwater harvesting. The analysis also indicates that non-conventional WSs are now more prominent in the literature than conventional ones. Overall, the present study demonstrated that modern water management strategies increasingly emphasize optimization and circular reuse. In contrast, earlier approaches tend to focus more on water conservation and economic efficiency. The literature also indicates a gradual shift from traditional supply-dominant models toward integrated, cost-effective, and sustainability-oriented approaches that combine multiple sources and advanced allocation techniques. Full article

The Puebla Metropolitan Area, one of the most industrialized regions in Mexico, shows severe contamination of both surface and groundwater. In this study a multi-tracer approach combining hydrochemistry with environmental isotopes (δ²H, δ¹⁸O, ³H) was applied to evaluate groundwater–surface water (GW–SW) interactions and their role in water quality degradation. Elevated concentrations of aluminum, iron, zinc, and lead were detected in the Alseseca and Atoyac Rivers, exceeding national standards, while arsenic, manganese, and lead in groundwater surpassed Mexican and WHO drinking water limits. The main sources of contamination include volcanic inputs from Popocatepetl activity (e.g., arsenic) and untreated discharges from industrial parks (e.g., lead), which together introduce significant loads of Potentially Toxic Elements (PTEs) into surface and groundwater. Isotopic analysis identified three sources for aquifer recharge: (1) recharge from high-altitude meteoric water, (2) mixed GW–SW water recharged at intermediate elevations with heavy metal presence, and (3) recharge from lower altitudes (evaporate water). Tritium confirmed both modern and old recharge, while isotope-based mixing models indicated surface water contributions to groundwater ranging from 18% to 72%. These interpretations were derived from the integrated analysis of hydrochemical and isotopic data, allowing the quantification of recharge sources, residence times, and mixing processes. The results demonstrate that hydraulic connectivity, enhanced by fractures and faults, facilitates contaminant transfer from polluted rivers into the aquifer. Full article

Uncertainty quantification (UQ) is critical for predicting solute transport in heterogeneous porous media, with applications in groundwater management and contaminant remediation. Traditional UQ methods, such as Monte Carlo (MC) simulations, are computationally expensive and impractical for real-time decision-making. This study introduces a novel machine learning framework to address these limitations. We developed a surrogate model for a 2D advection-dispersion solute transport model using a Bayesian Neural Network (BNN). The BNN was trained on a synthetic dataset generated by simulating solute transport across various stochastic permeability and dispersivity fields. Uncertainty was quantified through variational inference, capturing both data-related (aleatoric) and model-related (epistemic) uncertainties. We evaluated the framework’s performance against traditional MC simulations. Our BNN model accurately predicts solute concentration distributions with a mean squared error (MSE) of 9.8 × ${10}^{-5}$, significantly outperforming other machine learning surrogates. The framework successfully quantifies uncertainty, providing calibrated confidence intervals that align closely with the spread of the MC results. The proposed approach achieved a 98.5% reduction in computational time compared to a standard Monte Carlo simulation with 1000 realizations, representing a 65-fold speed-up. A sensitivity analysis revealed that permeability field heterogeneity is the dominant source of uncertainty in plume migration. The developed machine learning framework offers a computationally efficient and robust alternative for quantifying uncertainty in solute transport models. By accurately predicting solute concentrations and their associated uncertainties, our approach can inform risk-based decision-making in environmental and hydrogeological applications. The method shows promise for scaling to more complex, three-dimensional systems. Full article

Predicting the quality of freshwater, both surface and groundwater, is essential for the sustainable management of water resources. This study collected 1822 articles from the Scopus database (2000–2024) and filtered them using Topic Modeling to create the study corpus. The B-SLR analysis identified exponential growth in scientific publications since 2020, indicating that this field has reached a stage of maturity. The results showed that the predominant techniques for predicting water quality, both for surface and groundwater, fall into three main categories: (i) ensemble models, with Bagging and Boosting representing 43.07% and 25.91%, respectively, particularly random forest (RF), light gradient boosting machine (LightGBM), and extreme gradient boosting (XGB), along with their optimized variants; (ii) deep neural networks such as long short-term memory (LSTM) and convolutional neural network (CNN), which excel at modeling complex temporal dynamics; and (iii) traditional algorithms like artificial neural network (ANN), support vector machines (SVMs), and decision tree (DT), which remain widely used. Current trends point towards the use of hybrid and explainable architectures, with increased application of interpretability techniques. Emerging approaches such as Generative Adversarial Network (GAN) and Group Method of Data Handling (GMDH) for data-scarce contexts, Transfer Learning for knowledge reuse, and Transformer architectures that outperform LSTM in time series prediction tasks were also identified. Furthermore, the most studied water bodies (e.g., rivers, aquifers) and the most commonly used water quality indicators (e.g., WQI, EWQI, dissolved oxygen, nitrates) were identified. The B-SLR and Topic Modeling methodology provided a more robust, reproducible, and comprehensive overview of AI/ML/DL models for freshwater quality prediction, facilitating the identification of thematic patterns and research opportunities. Full article

Arid regions in Central Asia face persistent and increasing water scarcity, with groundwater serving as the primary source for drinking water, irrigation, and industry. The effective exploration and management of groundwater resources are critical, but are constrained by limited monitoring infrastructure and complex hydrogeological settings. This study investigates the Akbakay aquifer, a representative area within Central Asia with challenging hydrogeological conditions, to delineate potential zones for fresh groundwater exploration. A multi-criteria decision analysis was conducted by integrating the Analytical Hierarchy Process (AHP) with Geographic Information Systems (GIS), supported by remote sensing datasets. To address the subjectivity of weight assignment, the AHP results were further validated using Monte Carlo simulations and fuzzy logic aggregation (Fuzzy Gamma). The integrated approach revealed stable high-suitability groundwater zones that consistently stand out across deterministic, probabilistic, and fuzzy assessments, thereby improving the reliability of the groundwater potential mapping. The findings demonstrate the applicability of combined AHP–GIS methods enhanced with uncertainty analysis for sustainable groundwater resource management in data-scarce arid regions of Central Asia. Full article

The quality of groundwater, a crucial freshwater resource in cold regions, directly affects human health. This study used groundwater quality monitoring data collected in the central Songnen Plain in 2014 and 2022 as a case study. The improved DRASTICL model was used to assess the vulnerability index, while water quality indicators were selected using a random forest algorithm and combined with the entropy-weighted groundwater quality index (E-GQI) approach to realize water quality assessment. Furthermore, self-organizing maps (SOM) were used for pollutant source analysis. Finally, the study identified the synergistic migration mechanism of NH₄⁺ and Cl⁻, as well as the activation trend of As in reducing environments. The uncertainty inherent to health risk assessment was considered by developing a kernel density estimation–trapezoidal fuzzy number–Monte Carlo simulation (KDE-TFN-MCSS) model that reduced the distribution mis-specification risks and high-risk misjudgment rates associated with conventional assessment methods. The results indicated that: (1) The water chemistry type in the study area was predominantly HCO₃⁻–Ca²⁺ with moderately to weakly alkaline water, and the primary and nitrogen pollution indicators were elevated, with the average NH₄⁺ concentration significantly increasing from 0.06 mg/L in 2014 to 1.26 mg/L in 2022, exceeding the Class III limit of 1.0 mg/L. (2) The groundwater quality in the central Songnen Plain was poor in 2014, comprising predominantly Classes IV and V; by 2022, it comprised mostly Classes I–IV following a banded distribution, but declined in some central and northern areas. (3) The results of the SOM analysis revealed that the principal hardness component shifted from Ca²⁺ in 2014 to Ca²⁺–Mg²⁺ synergy in 2022. Local high values of As and NH₄⁺ were determined to reflect geogenic origin and diffuse agricultural pollution, whereas the Cl⁻ distribution reflected the influence of de-icing agents and urbanization. (4) Through drinking water exposure, a deterministic evaluation conducted using the conventional four-step method indicated that the non-carcinogenic risk (HI) in the central and eastern areas significantly exceeded the threshold (HI > 1) in 2014, with the high-HI area expanding westward to the central and western regions in 2022; local areas in the north also exhibited carcinogenic risk (CR) values exceeding the threshold (CR > 0.0001). The results of a probabilistic evaluation conducted using the proposed simulation model indicated that, except for children’s CR in 2022, both HI and CR exceeded acceptable thresholds with 95% probability. Therefore, the proposed assessment method can provide a basis for improved groundwater pollution zoning and control decisions in cold regions. Full article

This paper revisits the theoretical framework of computing the stabilization value of groundwater in a dynamic context. Specifically, we argue that what the existing studies have measured as the augmentation value contains a different type of value, the dynamic reallocation value (DRV), which should be evaluated as an important part of the dynamic stabilization value. We examine the existence of DRV and its underlying behavioural mechanism using a simple two-stage model and then generalise the specification to a dynamic model with an arbitrary number of stages. We find that behind the positive values of DRV, users amplify their reactions against surface water fluctuations and still realize a higher total expected benefit than in the case without uncertainty. Ignoring the existence of DRV may impair the value of groundwater as an essential instrument for climate adaptation. Full article

Understanding the groundwater movement patterns and regulating functions of the karst vadose zone is essential in addressing water scarcity and protecting the ecological environment in the karst area of southwest China. A laboratory-scale experimental model of a typical karst vadose zone was constructed and used to simulate the water flow process under the influence of four factors: transfer zone thickness, surface slope, karstification degree, and rainfall intensity. A corresponding distributed model was subsequently developed to simulate the laboratory experiments. The discharge recession process, the regulating capacity, and the division of fast and slow flow were quantitatively analyzed by the recession coefficient, the regulating coefficient, and the percentage of fast flow and that of sinkhole flow. As the transfer zone thickness increases from 40 cm to 120 cm, the vadose zone regulating coefficient rises from 0.49 to 0.53, while fast flow decreases from 87.7% to 78.1%, indicating that the enhanced regulating capacity is mainly governed by the slow flow system. The evident difference in growth rates between the percentage of fast flow (an increase of 9.1%) and that of sinkhole flow (an increase of 48.7%) indicates that the decrease in regulating capacity resulting from an increase in surface slope is primarily due to enhanced water loss through sinkholes. When the structure of the karst vadose zone remains constant, the regulating coefficient decreases exponentially with increasing rainfall intensity and gradually approaches a constant value, which represents the maximum regulating capacity of the karst vadose zone under its current structural conditions. Full article

To investigate differences in water and salt transport during irrigation, freezing, and thawing periods in typical saline-affected paddy fields and saline-affected upland fields, field-based automated in situ monitoring was conducted in both types of saline-affected farmland (May 2023 to May 2024). Correlation analysis identified seasonal drivers of water–salt migration, while the HYDRUS-3D model simulated transport and equilibrium processes. The HYDRUS-3D model, equipped with a freeze–thaw module, accurately simulated complex water–salt transport in cold arid regions. Key findings include: (1) During freeze–thaw periods, soil moisture content and electrical conductivity (Ec) increased with the retreating frost front in both upland and paddy soils. During the irrigation period, maximum soil moisture content and Ec values occurred at 80 cm depth in dryland soils and 60 cm depth in paddy soils, primarily influenced by irrigation and capillary rise. (2) Groundwater salt ions significantly affected soil salinization in both farmland types. During the freeze–thaw period, Ec positively correlated with soil temperature. During the irrigation period, Ec positively correlated with evapotranspiration and negatively correlated with precipitation. (3) Salt changes during the irrigation, freezing, and thawing periods were −565.4, 326.85, and 376.55 kg/ha for upland fields, respectively; corresponding changes for paddy fields were −1217.0, 280.07, and 299.35 kg/ha. (4) Both land types exhibited reduced salinity during the irrigation period, with paddy fields showing a reduction 3.36 times greater than dryland fields. During the freezing and thawing periods, both land types experienced salinity accumulation, with dryland fields accumulating higher salinity levels than paddy fields. These results indicate that paddy field irrigation and drainage systems help mitigate salinization, while dryland fields are more prone to springtime salt accumulation. These findings provide a basis for developing targeted management strategies for saline–alkali soils. Full article

In deep excavation dewatering engineering, fully enclosed cutoff walls are widely implemented to improve the efficiency of dewatering in the pit and prevent adverse environmental impacts such as land subsidence and damage to adjacent infrastructure. However, the presence of such impermeable barriers fundamentally alters flow dynamics, rendering conventional aquifer test interpretation methods inadequate. This study presents a novel closed-form analytical solution for transient drawdown in a leaky confined aquifer bounded by a rectangular, fully enclosed cutoff wall under constant-rate pumping. The solution is rigorously derived by applying the mirror image method within a superposition framework, explicitly accounting for the barrier effect of the curtain. A type-curve matching methodology is developed to inversely estimate key aquifer parameters—transmissivity, storativity, and vertical leakage coefficient—while incorporating the geometric and boundary effects of the curtain. The approach is validated against field data from a pumping test conducted at a deep excavation site in Wuhan, China. Excellent agreement is observed between predicted and measured drawdowns across multiple observation points, confirming the model’s fidelity. The proposed solution and parameter estimation technique provide a physically consistent, analytically tractable, and computationally efficient framework for interpreting pumping tests in constrained aquifer systems, thereby improving predictive reliability in dewatering design and supporting sustainable groundwater management in urban underground construction. Full article