Agricultural irrigation systems can bear the heavy losses of water through runoff, evaporation, and deep percolation. Crop water requirements vary with crop species, growth stage, soil properties, evapotranspiration, weather, and plant water status; therefore, irrigation scheduling based on a single parameter is insufficient. Conventional water management systems often rely on uniform application, fixed scheduling, and limited real-time information, which overlooks spatial and temporal variation in soil, crop, and weather conditions and can result in water loss, over-irrigation, and under-irrigation. This review highlights studies on the integration of AI-driven sensing and imaging technologies, IoT-based precision irrigation, variable-rate technologies, digital twins, and agricultural robotics for crop water management. These approaches integrate field, crop, weather, and spatial information to support water-demand assessment, crop-stress monitoring, irrigation scheduling, site-specific application, and automated water management. The reviewed studies report water savings, improved water-use efficiency, crop performance, and more site-specific irrigation and input application across different crops and irrigation systems. The review also highlights the limitations of these systems, including sensing reliability, model transferability, connectivity, interoperability, cybersecurity, investment costs, and environmental effects. Overall, the integration of these technologies provides potential for timely and field-specific irrigation decisions and supports efficient and climate-resilient water management in agriculture.
Norm-based nudges are increasingly used to promote environmentally responsible behavior, including more sustainable water use. It is widely recognized that such interventions should be informed by prior survey-based research. A more specific and often neglected question, however, concerns what those surveys should actually measure. This article addresses that question in the case of household tap-water use. We argue, first, that social norms often influence behavior not only directly but also indirectly, through mediating constructs such as personal norms, attitudes toward the behavior, and perceived behavioral control. Second, drawing on the Reasoned Action Approach and related work on mediated norm effects, we argue that these constructs are grounded in underlying behavioral, normative, and control beliefs. Accordingly, surveys designed to inform norm-based interventions should not focus primarily on self-reported behavior alone, but should also identify the belief patterns through which normative influence is likely to operate in a given population. On this basis, we propose a belief-sensitive survey blueprint for household tap-water use. The paper's contribution is therefore not to advance a new general theory of norm nudging, but to clarify how an indirect-pathways perspective can inform the design of more context-sensitive and belief-sensitive water interventions.
Nibras M. Mahdi, Jenan S. Sherza, Zainab T. Al-Sharify, Amar S. Abdul-Zahra, Luttfi A. Al-Haddad, Ahmed Ali Farhan Ogaili
Flow distributors are important components in water-flow and compact thermal-fluid systems because non-uniform flow division among parallel channels can disturb pressure balance and convective heat-transfer stability. However, evaluating distributor performance through repeated experimental trials and Computational Fluid Dynamics (CFD) simulations can be time-consuming during early-stage design. This work develops an Artificial Intelligence (AI)-assisted framework for predicting and classifying flow maldistribution in 3D-printed water-flow distributors using combined experimental and CFD-supported data. The dataset was obtained from water-flow measurements of rectangular distributors, conical distributors with an obstacle, and conical distributors without obstacle, while the CFD-supported dataset included Volume of Fluid (VOF) and single-phase flow results for selected conical configurations. Distributor family, height, flow rate, data source, and simulation method were used as input variables, while the non-uniformity flow coefficient, Φ, was used as the prediction target. An Extreme Gradient Boosting (XGBoost) regression model was developed to estimate Φ, and a classification layer was used to assign maldistribution severity levels. The model achieved coefficient of determination (R 2) = 0.962, Mean Absolute Error (MAE) = 0.126, Root Mean Square Error (RMSE) = 0.161, Mean Absolute Percentage Error (MAPE) = 4.87%, and Coefficient of Variation of Root Mean Square Error (CVRMSE) = 6.12%. The model achieved accuracy = 92.22%, precision = 92.90%, recall = 92.22%, and F1-score = 92.36%. The results showed that conical distributors generally produced lower maldistribution than rectangular distributors, while the AI model accurately captured both numerical Φ behavior and severity trends. The framework provides a tool to screen distributor designs and support CFD-guided water-flow and thermal-fluid system development.
Agricultural water use in inland river basins of northwestern China is affected by multiple interacting factors, and the associated indicators often show strong collinearity. To address these issues, this study developed a hybrid prediction model based on principal component analysis (PCA) and Lasso regression to establish an interpretable agricultural water-use prediction framework for arid inland river basins. Twenty indicators describing meteorology, water-use structure, water resources, and socioeconomic conditions in the Heihe River Basin from 2001 to 2023 were used for correlation analysis and PCA. An expanding-window time-series cross-validation scheme was then applied to determine the Lasso regularization parameter, and model performance was evaluated using R 2, root mean squared error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE). Agricultural water use ranged from 24.01 × 108 to 32.20 × 108 m 3, accounting for 77.26–85.72% of total water supply. It declined by 21.20% from 2001 to 2012, rebounded markedly in 2013, and generally decreased after 2014. Correlation analysis showed that several groups of indicators had correlation coefficients above 0.80. Eight principal components explained 97.70% of the total variance, from which 14 input indicators were retained. The optimal λ was 0.02. The PCA-Lasso model achieved an R 2 of 0.997, with RMSE, MAE, and MAPE values of 0.152, 0.114, and 0.495%, respectively. The influence of the variables related to water-resource and water-use structure was the most significant. These results indicate that the PCA-Lasso model can effectively simulate and predict agricultural water use in the Heihe River Basin and can support agricultural water management in arid regions.
Bennet Benjamin S, Sumanth Ramamoorthy Rathnasekaran, Rohit Anand Patnaik, Sasithradevi Anbalagan, Vijayalakshmi M
Detection of underwater waste in freshwater environments remains a challenging task due to the scarcity of domain-specific annotated datasets and adverse visual conditions caused by poor visibility, inconsistent illumination, color degradation, and suspended particles. Unlike existing underwater debris datasets that primarily focus on marine environments, freshwater pond ecosystems remain relatively underrepresented. We introduce PondDebris, a publicly available underwater waste detection dataset comprising 1,526 images and 3,224 bounding box annotations across two ecologically relevant categories: biodegradable and non-biodegradable waste. The dataset was collected under controlled aquarium conditions designed to emulate several visual characteristics of shallow freshwater ponds. Furthermore, we propose UWAD-Net (UnderWater Attention Debris Network), an attention-enhanced object detection framework that integrates the Convolutional Block Attention Module (CBAM) into YOLOv8s to improve feature representation and localization of underwater waste. YOLOv8s provides a lightweight and efficient detection backbone, while CBAM enhances channel and spatial feature attention for improved feature discrimination in challenging underwater scenes. Experimental results demonstrate that UWAD-Net achieves an mAP50 of 0.7294, mAP50-95 of 0.4517, precision of 0.8206, and recall of 0.6421. Cross-dataset evaluation on an external underwater waste dataset yielded an mAP50 of 0.6629 and a precision of 0.7738, demonstrating its generalization capability. The proposed dataset and detection framework provide an effective solution for freshwater underwater waste monitoring and support the development of intelligent environmental management systems.
