Groundwater serves as a lifeline in Egypt’s hyper-arid Eastern Desert, particularly for agricultural and domestic uses. However, a comprehensive understanding of groundwater origin, quality, and recharge dynamics in the region remains limited due to geological complexity, data scarcity, and the high cost of isotopic analysis. This study addresses these challenges by integrating stable isotopes (δ¹⁸O and δ²H), hydrogeochemical parameters, remote sensing, and explainable artificial intelligence (AI) to investigate groundwater dynamics and support sustainable water management strategies. A total of 34 groundwater samples were collected from three key aquifers: the Quaternary alluvium, Nubian Sandstone, and fractured Basement aquifers. Hydrochemical analyses and isotopic signatures distinguish meteoric water from paleowater sources, revealing significant mixing and recharge processes. The findings indicate that the Quaternary aquifer is increasingly influenced by upward leakage from the Nubian aquifer, facilitated by deep-seated faults. Between 2014 and 2021, water levels in the Quaternary aquifer declined by up to 14 m due to over-extraction, particularly in agricultural zones. To enhance predictive capabilities, a Support Vector Machine (SVM) model was developed to estimate δ¹⁸O values using multiple hydrochemical indicators, achieving strong performance (R² = 0.92, MSE = 2.89). SHapley Additive exPlanations (SHAP) analysis identified Mg²⁺, HCO₃⁻, and SO₄²⁻ as dominant factors influencing isotope variation. This integrated approach represents a novel application of explainable machine learning in hydrogeology and offers a scalable, cost-effective tool for assessing groundwater systems in arid regions. The study contributes directly to national water security goals and supports the global Sustainable Development Goal 6 (SDG 6) for clean water and sanitation.

Accurate and timely crop yield estimation is essential for effective agricultural management and global food security, particularly for winter wheat. This study aimed to assess the effectiveness of EnMAP hyperspectral imagery in combination with machine learning and deep learning models for winter wheat yield prediction in Hungary. Using EnMAP images from February and May 2023, along with ground truth yield data from four fields, we derived 10 distinct vegetation indices. Random Forest, Gradient Boosting, and Multilayer Perceptron algorithms were employed, and model performance was evaluated using Mean Absolute Error (MAE) and Coefficient of Determination (R²) values. The results consistently demonstrated that integrating multi-temporal data significantly enhanced predictive accuracy, with the MLP model achieving an R² of 0.79 and an MAE of 0.27, notably outperforming single-date predictions. Shortwave infrared (SWIR) indices were particularly critical for early-season yield estimations. This research highlights the substantial potential of hyperspectral data and advanced machine learning techniques in precision agriculture, emphasizing the promising role of future missions such as CHIME in further refining and expanding yield estimation capabilities.