In arid regions such as the Jordan valley, salinity and sodicity are major constraints to soil quality and crop production. Accurate spatial determination of sodicity and salinity at field scale is a challenge, which can limit the effectiveness of management strategies. Interpolation techniques are used to derive maps to estimate the extent of the areas affected by sodicity and devise appropriate management plans. Nevertheless, different methods may draw different pictures. The main objectives of this study are to compare two interpolation techniques: 1. empirical Bayesian (EBK) and 2. disjunctive kriging (DK) to spatially predict soil salinity and sodicity in intensively used agricultural soils. Surface and subsurface samples were collected from randomly selected agricultural fields and analyzed for salinity (ECe) and sodicity (sodium adsorption ratio -SARe and exchangeable sodium percentage -ESP). Both EBK and DK methods revealed serious soil salinization and sodification problems in the middle and southern parts of the Jordan Valley. Salinity (ECe) maps showed that about 34% of the total area has salinity < 4, 12% < 8, 7% < 16, and 46% exceeds 16 dS m−1. For sodicity (ESP), 44% < 10, 18% < 15, and 37% > 15. Surface soils had higher salinity and sodicity levels than subsurface soils. The average values of surface soils were ECe (15.7 dS m−1), SARe (9.8), and ESP (15.5), compared with ECe (7.4 dS m−1), SARe (7.5), and ESP (13.1) for subsurface soils. Smoother and less patchy predictions were generated using DK compared to EBK. However, EBK had higher accuracy than DK in spatially predicting and addressing the uncertainty inherent in soil salinity and sodicity. This investigation gives important fundamental steps for developing site-specific reclamation techniques to manage and sustain agriculture in these regions.