Estimation of the Electrical Conductivity of Saturated Paste Extract Using a Dielectric Sensor

The electrical conductivity of saturated soil-paste extract (${\mathrm{EC}}_e$) has been extensively used as the standard index for soil salinity assessment. However, this method is time-consuming and laborious. A number of research works are therefore attempting to find conceptual or empirical models to correlate ${\mathrm{EC}}_e$ with the soil bulk electrical conductivity ${\mathrm{EC}}_a$, which could be more easily provided in different ways. Dielectric devices, which could provide measurements of the soil dielectric constant and of ${\mathrm{EC}}_a$ simultaneously and for the same volume of soil, could be easily used to provide measurements of the bulk saturated paste electrical conductivity (${\mathrm{EC}}_p$). In the present work, a less time-consuming method of calculating ${\mathrm{EC}}_e$ from the bulk electrical conductivity of saturated paste (${\mathrm{EC}}_p$), as this can be obtained directly by a dielectric sensor (e.g., WET sensor), is examined. The method utilizes a model with two empirical relationships that estimate the solid particle surface charge component of electrical conductivity (${\mathrm{EC}}_s$) and the water content of the continuous liquid current flow-coupling pathway (${\theta }_{wc}$) as functions of the saturation percentage (SP). The appropriateness of the method was tested in 220 soil samples, collected from five different locations across Greece. The model was applied with two different couples of empirical relationships, one developed for U.S. soils and the other for Greek soils. The ${\mathrm{EC}}_e$ prediction in both cases was reliable and satisfactory (RMSE = 1.22 and $1.36\mathrm{dS}·{\mathrm{m}}^{-1}$). More specifically the results showed that in three out of five locations, the performance of the model supported by the empirical relationships for Greek soils was relatively better. In one location with special characteristics, the model with empirical relationships for U.S. soils performed better.