Search Results (3,479)

This study aims to explore a leakage-aware and explainable machine learning framework for predicting tunnel water inrush quantity (WIQ) under small-sample and high-heterogeneity geological conditions. A project-level dataset was compiled at a fixed spatial granularity of 30 m per excavation segment by integrating forward prospecting outputs, construction-face observations, and geological reports, and six hydrogeological–structural indicators were used to predict the water inflow rate in cubic meters per hour. To overcome data scarcity and improve generalization, a tabular generative adversarial network (GAN) was introduced to augment the training distribution while preserving marginal statistics and inter-variable dependence, and a swarm-intelligence optimizer was employed to tune a Categorical Boosting (CatBoost) regressor for stable performance. In addition, six mainstream tree-based learners were benchmarked under a unified protocol, and model transparency was ensured through a multi-level interpretability suite combining SHapley Additive exPlanations (SHAP) attribution, partial dependence with individual conditional expectation (ICE) diagnostics, and interaction surfaces. Results show that, under the present fixed split, training-set augmentation was associated with improved performance for the evaluated baseline learners, and the proposed hybrid model achieved encouraging hold-out accuracy. However, because the dataset contains only 55 real samples and the test set contains only 11 real samples, the reported performance should be interpreted as an initial project-specific indication rather than robust evidence of generalizable reliability. Interpretability analyses further identify lithologic and reflector-related factors as dominant drivers, and reveal nonlinear response patterns and interaction-sensitive high-risk regions. Overall, the proposed framework shows potential to improve predictive performance and engineering interpretability for the studied project, and may provide a useful reference for drainage and reinforcement planning. Further confirmation through repeated data splitting, additional samples, and external validation is still needed before broader application. Full article

Water scarcity has become a major challenge for agriculture, particularly in arid and semi-arid regions where irrigation is essential for sustaining crop and forage production. As freshwater supplies face growing pressure from climate change, urban growth, and industrial use, there is increasing interest in exploring alternative water sources to support sustainable agriculture. Produced water, a byproduct of oil and gas extraction, may represent an alternative water source in water-limited regions like the southwestern United States and the Middle East. However, raw produced water often contains high levels of salinity, trace metals, hydrocarbons, and naturally occurring radioactive materials, which cause risks to soils, crops, livestock, and food systems. This review synthesizes peer-reviewed studies up to January 2026 and reports on the agricultural application of treated produced water, focusing on its effects on soil properties, crop growth, yield, and forage nutritive quality. Existing research shows that treated produced water could be used for grain as well as forage crops under controlled conditions, but poorly treated and managed applications can lead to increases in soil salinity, structural degradation, reduced nutrient uptake, and hindered crop performance. In forage systems, irrigation with treated produced water has also been associated with changes in nutritive value, increasing concerns for livestock health. Several knowledge gaps remain, including limited long-term field studies, insufficient information on crop-specific contaminant thresholds, incomplete assessment of treatment and remediation strategies under different environmental conditions, and the absence of a consistent framework for classifying the chemistry of treated produced water for agricultural applications. Addressing these gaps through integrated soil, crop, and water research and the development of clear policies and guidelines is essential for determining whether treated produced water can be safely and sustainably used in agriculture under growing water scarcity. Full article

The reuse of treated oilfield-produced water (PW) presents a viable solution to water scarcity in arid regions; however, elevated boron (B) levels pose a significant constraint for sustainable irrigation. This study evaluates boron dynamics in a soil–plant system irrigated with treated PW and examines the effectiveness of nature-based solutions in mitigating its accumulation. A controlled experiment using two soil types and multiple water sources was conducted, with biochar and gypsum applied as soil amendments. Boron concentrations were assessed in plant tissues, roots, and soil layers. Results showed significant boron accumulation under PW irrigation, exceeding safe agronomic thresholds, and soil analysis indicated greater boron retention in surface layers. Boron concentrations reached maximum average concentrations exceeding 200 mg kg⁻¹. To elucidate species-specific tolerance mechanisms, bioaccumulation factors (BAFs) and translocation factors (TFs) were calculated. Results revealed a distinct root-trapping strategy, with high BAF values under oilfield-produced water, while TF values remained significantly lower, indicating that these forage species successfully restricted boron translocation to aerial tissues. Full article

Groundwater is the primary source of livestock watering across the arid pasturelands of western Kazakhstan, yet no systematic field hydrochemical assessment has been published for this region in over 40 years. This study presents the first systematic field-based hydrochemical characterisation of groundwater sources used for pasture livestock watering in the West Kazakhstan Region and Aktobe Region, filling a critical data gap that has persisted since the Soviet era. Specifically, it characterises the hydrochemistry, water quality, and infrastructure condition of groundwater sources, and evaluates the groundwater resource potential against current and projected livestock water demand. A total of 139 groundwater samples were collected along 11,182 km of field routes during May–July 2025, and analysed for 25 physicochemical parameters; hydrochemical classification was performed using AquaChem 11, and spatial analysis was conducted in ArcGIS 10.8. The groundwater chemistry distribution is bimodal: fresh bicarbonate-calcium-magnesium waters (TDS < 3.0 g/L) constitute approximately 80% of samples, while highly mineralised chloride-sulphate-sodium waters (TDS up to 9.91 g/L) occur in salt-dome-influenced discharge zones. Nitrate concentrations exceeded 50 mg/L in 23–36% of samples, with maxima of 635 mg/L, reflecting intensive anthropogenic contamination near livestock facilities. Predictive exploitable fresh groundwater resources exceed current livestock demand by a factor of 162. The principal constraint on pasture water supply is not resource scarcity but the non-operational status of 51–75% of inspected watering infrastructure, a legacy of post-Soviet institutional collapse that requires urgent rehabilitation. Full article

Winter wheat (Triticum aestivum L.) production in the Yellow River Delta is limited by saline–alkali soils and freshwater scarcity, while the responses of different straw-returning modes under contrasting irrigation regimes remain unclear. A field experiment was conducted with two irrigation regimes, normal irrigation (W1) and deficit irrigation (W2), and four straw-returning modes, direct straw return (RS), straw-derived cattle manure return (RM), straw biochar return (RB), and straw pellet return (RG). The experiment followed a split-plot randomized block design with three replicates. Soil properties, leaf physiology, photosynthetic performance, grain yield, and irrigation water use efficiency (IWUE) were evaluated. Compared with W2, W1 increased mean grain yield by 9.4%, whereas W2 increased mean IWUE by 36.7%. Among the straw-returning modes, RS showed the most consistent performance. Under W1, W1RS produced the highest grain yield (3509.72 kg ha⁻¹). The stable performance of RS was characterized by relatively favorable soil moisture status, lower MDA content, higher antioxidant enzyme activity, and better maintenance of Pn. Pearson correlation analysis showed that grain yield was positively correlated with Pn and CAT activity, whereas MDA was negatively correlated with Pn. These results suggest that RS may be a feasible straw-returning mode for winter wheat production in saline–alkali soil. Full article

Global warming and associated environmental changes are reducing arable land and intensifying salinization risks, posing growing threats to food security. Soil salinity is an increasing threat to agricultural productivity worldwide, particularly in arid and semi-arid areas. Wheat (Triticum aestivum L.) is one of the most important and widely cultivated cereal crops for human consumption and livestock feed. However, with increasing water scarcity and the incidence of salt-affected lands, wheat productivity is increasingly affected by salinity. Previous studies have investigated salinity tolerance mechanisms mainly at the seedling and reproductive stages of wheat; however, comparatively fewer studies integrate rapid biochemical and physiological responses during the first hours of germination stress exposure together with transcriptional analyses during early seedling establishment, even though this stage is critical for stand establishment. Here, we evaluated early physiological and transcriptional responses of salt-tolerant, moderate, and sensitive wheat cultivars exposed to 0 or 150 mM NaCl during germination and the early seedling stage. Tolerant and sensitive cultivars showed contrasting germination performance under salinity. Physiological analysis showed that salt-tolerant cultivars exhibited higher proline accumulation and higher antioxidant enzyme activities (CAT, SOD, and GR), while maintaining lower MDA levels under salinity compared with sensitive cultivars. Notably, tolerant cultivars showed marked upregulation of TaHKT1;4, TaP5CS, TaMYB, and TaDHN genes associated with ion homeostasis, osmoprotectant metabolism, and stress-responsive regulation. These responses represent integrated early-stage biochemical, physiological, and transcriptional indicators of salinity responsiveness rather than direct predictors of final yield performance. Full article

Evapotranspiration (ET) is a fundamental process within the water cycle and the agricultural water balance, optimizing resource allocation, maintaining soil health, and enhancing ecosystem resilience to climate change. Because ET represents a primary consumptive use of irrigation on agricultural lands, enhancing water-use efficiency and sustainable water management requires accurate estimation of evapotranspiration to support long-term sustainability and productivity. This study offers an effective means to visualize spatial and temporal patterns of reference evapotranspiration (ETo) across various vegetation management practices. This study examined the impacts of agroforestry buffers (ABs), grass buffers (GBs), biofuel crops in an agroforestry watershed (BCa), and biofuel crops in a grass buffer watershed (BCg) on ETo, compared to a corn (Zea mays L.)–soybean (Glycine max L.) rotation (RC) for claypan soil in Northern Missouri, USA. The experimental watersheds were located at the Greenley Memorial Research Center, Missouri, USA. Campbell Scientific sensors and Photosynthetically Active Radiation (PAR) smart sensors were installed to measure net radiation, anemometers, humidity, and air temperature. All instruments were mounted on masts at a height of 2 m above ground level in crop, tree, grass, and biofuel areas. Measured meteorological data were recorded hourly from April to October during 2017 and 2018. Daily ETo predictions were calculated using the Penman–Monteith model. These ETo predictions were displayed across the landscape using Python-based GIS for selected dates (each Saturday) for the watersheds. The methodology was implemented using the software programs of Python 2.7.10 and ArcGIS 10.3.1. The results indicated that ETo increased by 11%, 17%, 18%, and 25% in 2017, and by 7%, 9%, 14%, and 20% in 2018 for AB, BCa, BCg, and GB, respectively, compared to RC management. This process may improve soil water recharge in perennial management systems. Accurate estimation of ET in agricultural regions is critical for understanding water balance, hydrological and ecosystem processes, and climate variability. Given that agriculture constitutes the majority of global water consumption, precise ET estimation is particularly significant for sustainable water management, especially in regions experiencing water scarcity. These outcomes may support effective planning and management of agricultural water resources by enabling optimized irrigation and agricultural production. Full article

As a critical component of ice-snow tourism, land cover classification for ski resorts is crucial to ice-snow resource management. However, there is currently a scarcity of datasets and methods capable of high-precision mapping for such fine-grained scenarios. Although Transformers with long-sequence interactions and convolutional neural networks (CNNs) have emerged as mainstream solutions, their performance remains limited on high-resolution remote sensing data characterized by small datasets and high heterogeneity. Targeting land cover classification in ski resort areas, this study proposes a triple-branch segmentation framework integrating CNNs and Transformers to extract global, detail and boundary features (GDBNet), and constructs the first high-resolution ski resort land cover dataset with a resolution of 0.75 m using JiLin-1 satellite constellation (LULC_SKI). The framework employs a backbone combining SegFormer with dual CNN branches. SegFormer captures global semantic context, while dual ResNet-18 branches extract local semantics and edge details respectively. The neck integrates two specialized feature interaction modules, the proposed Pixel-Guided Feature Attention (PG-AFM) and Boundary-Guided Feature Attention (BG-AFM), which synergistically fuse these heterogeneous feature representations for enhanced multi-scale modeling. For the segmentation head, a multi-task learning approach supervises both semantic and edge outputs. LULC_SKI covers seven representative ski resorts in Jilin Province, China, comprising 10,000 multi-seasonal images annotated with six land cover classes, including roads, vegetation, built-up areas, ski runs, water bodies, and cropland. Experiments demonstrate GDBNet achieves 85.44% mIoU and 91.84% mF1 on LULC_SKI, outperforming other advanced models with particularly significant improvements for linear objects like roads and ski runs. Extensive experimental comparisons show that GDBNet delivers consistently excellent performance on both the iSAID and LoveDA datasets, underscoring the superiority of our proposed method. Ablation studies validate the effectiveness of the triple-branch architecture, attention modules, and multi-task supervision. This work proposes a modular framework for land cover classification in complex ski resort scenarios. Full article

As urban areas expand, the sustainable management of municipal water becomes a critical challenge, especially in arid and semi-arid regions facing severe water scarcity. Accurate assessment of urban plant water requirements (PWR) is essential for developing sustainable landscape architecture and resilient green infrastructure. In this study, a new quantitative equation (PWR_q) was developed as a regional proof of concept to adjust reference evapotranspiration estimates for hyper-arid conditions. A Tree Morphology Coefficient (K_tm) is introduced to combine canopy features (form, height) and leaf traits (size, density) with an updated drought-resistance coefficient (K_dr). Field measurements of 277 mature trees, representing 27 native and introduced species in Riyadh and Jeddah, Saudi Arabia, were analyzed. The framework explicitly includes an empirical multiplier to account for extreme urban heat island (UHI) effects and aerodynamic canopy scaling. Instead of direct empirical validation, the PWR_q model was benchmarked against established reference indices: Water Use Classification of Landscape Species (WUCOLS) and Simplified Landscape Irrigation Demand Estimation (SLIDE), showing strong alignment with established categorical indices and structural traits. The results confirm that the morphology-based method effectively makes previously subjective classifications objective. Notably, the quantitative assessment found that the dominant introduced species require about 3.5 times more water than native species. As a proof of concept, future research should empirically validate these findings against direct physical measurements, such as sap flow sensors or lysimeters. The proposed framework presents a practical, objective decision-support tool for municipal policymakers and landscape architects to optimize species selection, implement nature-based solutions (NBS), and achieve long-term sustainability in urban greening. Full article

In Morocco, snow constitutes a crucial freshwater resource, particularly in the Atlas Mountains, where seasonal snowpack significantly contributes to surface water availability, groundwater recharge, and down-stream water supply. However, snow monitoring in these regions remains challenging due to the scarcity and uneven distribution of ground-based snow depth measurements, especially at high altitudes. This lack of observations limits the accurate assessment of snowpack dynamics and hampers hydrological modeling and water resource management. In this study, we assessed the performance of an empirical approach to estimate snow depth from satellite-derived fractional snow cover (FSC) obtained from MODIS observations. Five empirical FSC snow depth models, including linear and nonlinear exponential formulations, are developed and applied across multiple regions of the Moroccan Atlas Mountains. Model coefficients are calibrated independently for each region using three complementary optimization techniques, nonlinear least squares regression, genetic algorithms, and simulated annealing. Model skill was evaluated during calibration and validation using the Kling–Gupta Efficiency (KGE), Pearson correlation coefficient (R), and absolute error metrics (RMSE and MAE). Results show substantial performance differences across formulations and regions. The most flexible exponential model achieved highest efficiency (KGE up to 0.87; R > 0.85) and 0.26 cm (MAE) under moderate snow conditions. Linear formulations exhibited limited robustness, whereas exponential models better captured snow depth dynamics, particularly in high-altitude areas with deep and persistent snowpacks. These results highlight the potential of FSC-based empirical modeling as a practical and operational solution for snow depth estimation in data-scarce mountainous regions of Morocco. Full article

Wastewater membrane bioreactors (MBRs) have become an important advanced treatment technology due to their ability to produce high-quality effluent suitable for discharge and water reuse. However, their broader and more sustainable application remains constrained by membrane fouling, elevated energy demand, and the operational complexity of coupled biological and membrane separation processes. This comprehensive review critically evaluates the growing application of machine learning (ML), explainable artificial intelligence (XAI), and digital twin (DT) technologies in MBR systems. Published studies on fouling prediction, energy optimization, effluent quality estimation, and intelligent operational support are critically evaluated, with explicit attention to model performance, dataset limitations, and generalizability. The reviewed literature shows that ML models, particularly ensemble methods, support vector machines, and deep learning approaches, have demonstrated strong potential for predicting major MBR performance indicators, including transmembrane pressure, permeate flux, fouling resistance, and selected effluent-quality variables. In parallel, XAI methods such as SHAP, LIME, and Anchors are increasingly being used to enhance model transparency and to reveal the dominant factors controlling process performance. Digital twin frameworks further extend this potential by enabling the integration of mechanistic understanding, online sensor data, data-driven prediction, and interpretable decision support within real-time operational platforms. Nevertheless, several barriers continue to hinder practical implementation, including the limited number of full-scale studies, the scarcity of openly accessible and standardized datasets, insufficient consideration of uncertainty and model drift, and the early-stage maturity of DT deployment in operational plants. The evidence reviewed suggests that integrating ML, XAI, and DT can substantially improve the reliability, interpretability, and operational efficiency of MBR systems. Future research should therefore focus on full-scale validation, the development of benchmark datasets, uncertainty-aware modeling, and practical deployment strategies for interpretable intelligent MBR management. Full article

Water security is crucial for human well-being and environmental sustainability. The rapid increase in urbanization, climate change, pollution, etc., leads to water scarcity in many parts of the world. Therefore, it is important to understand the concept and growing challenges of water security. Using the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) framework, 146 articles were identified for this study. The results highlight a novel aspect of the definition of water security, presenting it in a simpler, broader way with key components. The indicators for assessing water security are categorized into quantitative, qualitative, and combined types, and are further arranged across different dimensions, domains, and spatial scales. The study also examines Urban Water Security assessment methods and categorizes them into distinct methodological groups. Additionally, the studies show that only 25 articles explore artificial intelligence in the context of water security indicators. This reveals the need to address the gap between artificial intelligence and the assessment of water security. From these limited articles, artificial intelligence types and models were identified, and their applications were grouped into thematic categories. In general, this study supports improved assessment, decision-making, and sustainable water security management. Full article

Underwater image enhancement is essential for marine visual perception tasks. However, the highly heterogeneous optical degradations in real-world waters, the scarcity of paired training data, and the inherent dilemma for existing models in balancing long-range dependency modeling with computational overhead pose significant challenges. To address these issues, this paper proposes a prior-guided, self-supervised underwater image enhancement framework called Scene-Aware Retinex Mamba (SARM). This framework seamlessly integrates Retinex theoretical priors with state space models (SSMs) and operates without paired supervision by employing a prior-guided pseudo-labeling strategy to guide network optimization. Architecturally, SARM deeply couples the physical Retinex prior with SSM. Its core module integrates multi-color space features and leverages a 2D selective scan mechanism to achieve global context modeling with linear complexity $\mathcal{O}\left(HW\right)$, effectively removing complex color casts and suppressing non-uniform scattering noise. To further overcome the generalization bottlenecks in cross-domain underwater testing, this paper introduces a Scene-Aware Adapter (SAA), which facilitates dynamic loss scheduling and adaptive feature gating by quantifying scene-specific degradation characteristics. Comprehensive evaluations on multiple benchmark datasets, including UIEB, EUVP, and UCCS, demonstrate that SARM achieves state-of-the-art subjective and objective enhancement quality (e.g., yielding a URanker score of 2.491 and a CCF score of 35.76), while maintaining an ultra-fast inference speed of 136.52 FPS on the UIEB dataset. Furthermore, extended experiments reveal that SARM can significantly boost the performance of downstream vision tasks, validating its potential as a robust preprocessing module for various practical marine vision applications. Full article

In the face of intensifying climate change, agricultural water security in semi-arid zones has emerged as a critical frontier for water governance. This study provides a systematic and critical analysis of the scientific literature to map current research frontiers and structural gaps. The methodology integrated the PRISMA 2020 protocol and a modified Methodi Ordinatio, spanning a search period from 2014 to 2026 across the Science Direct and SciELO databases. From an initial broad screening, 136 high-impact articles were selected based on rigorous inclusion and exclusion criteria. The findings reveal a significant fragmentation of knowledge, characterized by a high prevalence of small-scale studies (25 articles) and limited interdisciplinarity. Notably, a governance-centric approach is present in only 20% of the literature, while the Water–Energy–Food Nexus appears in just 6%, signaling a major disconnect in holistic management. Based on these results, this study identifies water governance and socioeconomic integration as the most pressing research gaps. Consequently, an integrated conceptual framework is proposed, built upon three pillars: Governance, Technology, and Environment (GET). This study concludes that advancing the frontiers of agricultural water security requires moving beyond isolated solutions toward a structured, systemic, and interdisciplinary integration. Full article

Water scarcity in arid regions poses a significant challenge for the maintenance of Concentrated Solar Power (CSP) plants, where mirror cleaning consumes substantial resources. This study proposes a systematic methodological framework to bridge the gap between laboratory-scale surface characterization and engineering-scale water consumption. Through a gravimetric approach based on the physical principles of droplet retention (Furmidge theory), the water-saving potential of self-cleaning coatings has been quantified. Experimental results on 100 cm² specimens demonstrate that hydrophobic coatings can reduce residual water from 0.52 L/m² to approximately 0.24 L/m², achieving a water-saving potential of over 50%. The model incorporates a site-specific soiling factor (f_dirt) and was validated using field data from the ENEASHIP pilot plant. This approach provides a promising predictive tool for plant operators to optimize cleaning strategies and reduce operational costs, offering a scalable methodology for the solar thermal industry. Full article