International Journal of Plant & Soil Science

Diagnosis of soil salinity and its spatial variability is required to establish control measures in irrigated agriculture. Soil salinity can vary temporally and spatially due to dynamic nature of soluble salts. For these reasons, practical methods for measuring soil salinity (ECe) are required to achieve faster, cheaper, and reliable surveys. Thus, the bulk soil electrical conductivity (ECa) was measured directly in the field using portable sensors (EM₃₈) in vertical position (EM_V) or horizontal position (EM_H). Therefore, the objective of this research was to develop statistical relations between ECa and ECe and assess whether they could be applied for salinity predicting and mapping. The empirical relations were established to convert EM₃₈ readings (ECa) in mSm^-1 to ECe values in dSm^-1 for different depths. So, ECe was determined in soil samples collected from 33 sites at depths of 0-30, 30-60 and 60-90 cm. Also, 33 EM readings (EM_V and EM_H) taken at soil sampling sites were used to derive equations for salinity prediction. Fifteen quadratic or multiple regression models were established to describe the relations between the original or logarithmic values of both ECa and ECe. The statistical comparison measurements; coefficient of determination (R²), simple correlation coefficient (r) and t-test were also used to evaluate the relationships between the actual and predicted ECe values to select the relevant models. So, two preferable models were chosen and one equation for each soil depth was derived from each selected model to be used for ECe predicting. These equations allow useful prediction of ECe using ECa values, since ECa values are regressed strongly with ECe values (R² > 0.70). Also, the actual and the predicted ECe levels were highly correlated (r > 0.80), in addition to insignificant differences between them according to t-test.