This systematic review explores the use of digital twins (DT) for sustainable agricultural water management. DTs simulate real-time agricultural environments, enabling precise resource allocation, predictive maintenance, and scenario planning. AI enhances DT performance through machine learning (ML) and data-driven insights, optimizing water usage. In this study, from an initial pool of 48 papers retrieved from well-known databases such as Scopus and Web of Science, etc., a rigorous eligibility criterion was applied, narrowing the focus to 11 pertinent studies. This review highlights major disciplines where DT technology is being applied: hydroponics, aquaponics, vertical farming, and irrigation. Additionally, the literature identifies two key sub-applications within these disciplines: the simulation and prediction of water quality and soil water. This review also explores the types and maturity levels of DT technology and key concepts within these applications. Based on their current implementation, DTs in agriculture can be categorized into two functional types: monitoring DTs, which emphasize real-time response and environmental control, and predictive DTs, which enable proactive irrigation management through environmental forecasting. AI techniques used within the DT framework were also identified based on their applications. These findings underscore the transformative role that DT technology can play in enhancing efficiency and sustainability in agricultural water management. Despite technological advancements, challenges remain, including data integration, scalability, and cost barriers. Further studies should be conducted to explore these issues within practical farming environments.

Optimizing nitrogen (N) fertilization is essential for sustaining durum wheat yield and grain quality while reducing the environmental impacts associated with N over-application. Hyperspectral sensing provides a rapid and non-destructive approach for monitoring crop N status. However, high-dimensional data, phenology-dependent spectral responses, and spatial autocorrelation in field measurements limit robust nitrogen classification and interpretation. This study evaluated hyperspectral-based nitrogen status classification in durum wheat under Mediterranean field conditions and identified key spectral regions using explainable artificial intelligence. A field experiment was conducted in Southern Italy using ten N fertilization rates (0–180 kg N ha⁻¹). Canopy reflectance was acquired at the booting and heading stages from georeferenced sampling locations. Three nitrogen stratification strategies (binary Low–High, Extreme, and three-level) were evaluated using Random Forest, SVM-RBF, and XGBoost classifiers. Model performance was assessed using spatially independent Leave-One-Plot-Out cross-validation at both the sample and plot levels, with plot-level predictions derived through majority voting. Classification robustness was strongly influenced by the stratification strategy and phenological stage. The binary Low–High stratification achieved the highest sample-level accuracy, with a maximum of 0.78 at booting (SVM-RBF) and 0.75 at heading (SVM-RBF), whereas the Extreme stratification produced intermediate performance, with maximum accuracies of 0.73 at booting (SVM-RBF) and 0.63 at heading (XGBoost). Plot-level aggregation improved performance, reaching up to 0.90 at booting and 1.00 at heading. SHAP analysis highlighted red, red-edge, and near-infrared wavelengths as the dominant contributors, with increased reliance on longer wavelengths at the heading. Overall, explainable machine learning provides a robust framework for hyperspectral nitrogen monitoring in durum wheat.