Understanding surface-groundwater interactions is essential for sustainable water resource management in arid regions, where water scarcity and water quality degradation pose critical challenges. This study explored the hydrochemical characteristics, dominant controlling processes, and linkages between surface water and groundwater in the arid Middle Egypt basin. Hydrochemical facies and major physicochemical parameters were compared for both water types, supported by multivariate statistics (principal component analysis and PER-MANOVA) and machine-learning approaches (Random Forest and a hybrid K-nearest neighbor model). Results revealed distinct hydrochemical signatures, with surface waters characterized mainly by bicarbonate facies and groundwater exhibiting more mineralized sodium chloride and calcium chloride facies. Hydrochemical patterns suggest that surface water chemistry is influenced by evaporative concentration, whereas groundwater composition reflects mineral dissolution and cation exchange; however, these process interpretations are considered first-order and indicative rather than diagnostic. Random Forest analysis identified total cations, total anions, and electrical conductivity as the most important variables discriminating between surface water and groundwater. The hybrid KNN model indicated that surface water chemistry alone provided the strongest basis for predicting groundwater composition, with relatively high performance for Na, Cl, and Mg (R2 = 0.71-0.84) but weak performance for hardness and sulfate. Chemical similarity showed no systematic distance-decay relationship, and spatial proximity was not a primary control on surface-groundwater chemical relationships. These findings highlight contrasting hydrogeochemical controls on surface water and groundwater in the Nile Valley and demonstrate the value of integrating hydrochemical analysis with machine-learning tools for improving understanding of surface-groundwater interactions in arid environments.