This study assessed spatial variations in physical and chemical properties of an agricultural drain near Borg El-Arab city, Alexandria, Egypt. Pearson’s correlation coefficient indicated that salinity had strong correlations with total dissolved solids (TDS) (r 0.999, p < 0.001) and Cl− (r 0.807, p 0.016), whereas, pH was considerably affected by temperature (r 0.674, p 0.067), oxidation reduction potential (ORP) (r 0.866, p 0.006) and NO3− (r 0.731, p 0.039). Those results were further confirmed by applying an adaptive neuro-fuzzy inference system and regression models. Moreover, principal component analysis (PCA) indicated that PC1 explained 41.1% of the total variance, and had high loadings of TDS (0.46), salinity (0.46) and Cl− (0.48). Additionally, PC2 accounted for 35.2% of the total variance, and had high loadings of pH (0.53), temperature (0.48), ORP (0.40) and NO3− (0.47). The present study revealed that artificial intelligence and PCA could be used to effectively reduce the number of physicochemical parameters that may assist in the description of drainage water quality. It is recommended that the current status of the drain is suitable for reuse in irrigation purposes except at few locations containing high salinity.