Comparison of Soil EC Values from Methods Based on 1:1 and 1:5 Soil to Water Ratios and EC_e from Saturated Paste Extract Based Method

• Main Authors: George Kargas, Paraskevi Londra, Anastasia Sgoubopoulou
• Format: Article
• Language: English
• Published: MDPI AG 2020-04-01
• Series: Water
• Subjects: saturated soil paste; electrical conductivity; salinity
• Online Access: https://www.mdpi.com/2073-4441/12/4/1010
• ISSN: 2073-4441

The present study investigates the effect of three different methods of obtaining 1:1 and 1:5 soil-over-water mass ratios (soil:water) extracts for soil electrical conductivity (EC) measurements (EC_1:1, EC_1:5). On the same soil samples, also the electrical conductivity of the saturated paste extract (EC_e) was determined and the relationships between EC_e and each of the three of EC_1:1 and EC_1:5 values were examined. The soil samples used were collected from three areas over Greece (Laconia, Argolida and Kos) and had EC_e values ranging from 0.611 to 25.9 dS m⁻¹. From the results, it was shown that for soils with EC_e < 3 dS m⁻¹ the higher EC values were obtained by the method where the suspension remained at rest for 23 hours and then shaken mechanically for 1 h. On the contrary, no differences were observed among the three methods for soils with EC_e > 3 dS m⁻¹. Also, in the case of EC_1:5, the optimal times for equilibration were much longer when EC_e < 3 dS m⁻¹. Across all soils, the relationships between EC_e and each of three methods of obtaining EC_1:1 and EC_1:5 were strongly linear (0.953 < R² < 0.991 and 0.63 < RMSE < 1.27 dS m⁻¹). Taking into account the threshold of EC_e = 3 dS m⁻¹, different EC_e = f(EC_1:5) linear relationships were obtained. Although the linear model gave high values of R² and RMSE for EC_e < 3 dS m⁻¹, the quadratic model resulted in better R² and RMSE values for all methods examined. Correspondingly, in the 1:1 method, two of the three methods used exhibited similar slope values of the linear relationships independent of EC_e value (EC_e < 3 or EC_e > 3 dS m⁻¹), while one method (23 h rest and then shaken mechanically for 1 hour) showed significant differences in the slopes of the linear relationships between the two ranges of EC_e.