Electrical Conductivity and Sodium, Calcium and Magnesium Contents in Saturation Extract of Kamphaeng Saen Soil Series

Abstract

                  Several soil variables are routinely measured in the study of soil salinity. These variables are then used only for classifying soil as being saline. This study examined various relationships between all these variables to gain insight into the nature of salts presence in the field. The relationships will be useful for further study of salt distribution in the area. Agrid of 100 x 100 m² was laid out on the Kamphaeng Saen soil series (Typic Haplusalfs) in the field of 128 hectares in Kasetsart University, Kamphaeng Saen. Soil samples were taken for a total of 127 points, at the depth of 0-5 and 25-30 cm during April (summer) and November (end of rainy season) of 1989. Soil analysis included electrical conductivity of soil saturation extract (Ece), and of soil solution at the soil water ratio of 1:5(EC1:5), the total concentration of soluble salts: [Na], [Ca] and [Mg] in the soil saturation extract, pH of soil to water solution of 1:1, and soil water content on mass basis (θ_m). Calculated variables were sum of soluble salt content ([Na+Ca+Mg]) and practical sodium adsorption ratio (SAR_p = [Na]/([Ca+Mg])^0.5).

                    Salt concentrations and EC values were higher in the upper layer of surface soil. This was more pronounced in summer. However, salt concentrations became higher in the rainy season, indicating increasing solubility and movement by water. Basic statistical analysis showed that only pH and θ_m are normally distributed, with the arithmetic means of pH = 6.8 and θ_m = 14.21%. All other variables are log-normally distributed, with the log-means of ECe = 4.08 dS/m, [Na] = 7.84, [Ca] = 14.92, [Mg] = 6.80 and [Na+Ca+Mg] = 27.99 mol/m³ and SARp = 2.01 mol^0.5/m^1.5. The average ECe was on the borderling of being saline soil. The log-means have smaller magnitude than the arithmetic means.

                    Further analysis of the relationships between all nine variables showed that only 3 relationships are strongly correlated. These are the ralationships between ECe vs. EC1:5; ECe vs. soluble salt concentrations; and [Ca] vs. [Mg]. The correlation functions for the overall data (508 points) are as follows: i) ECe = 6.807EC1:5, (r = 0.956); ii) ECe = 0.8916 + 0.0731[Na] + 0.1702[Ca] + 0.0086(Mg), (r = 0.969) or [Na+Ca+Mg] =7.115ECe^0.9571, (r = 0.957); and iii) [Mg] = 0.3295[Ca]^1.024, (r = 0.882). Two other relationships have moderate r values. These are SARp = 0.9889 + 0.2716ECe, (r = 0.681); and pH = 6.27 + In([Na]^0.123[Ca]^0.368[Ng]^-0.449), (r = 0.398). The rest have low values of r (r<0.4), even though some relationships are of statistical significance.

                    The study leads to the conclusion that for the studied area, ECe can be evaluated from EC1:5; when EC1:5 is greater than 0.6 dS/m, the soil is saline. The dominant soluble salt is calcium. Calcium and magnesium behave similarly in their solubility. [Na] in the range of 3-5 mol/m³ is not related to [Ca] and [Mg], indicating a different solubility factor. At higher concentraiton, [Na] is positively related to [Mg]. When the total solube salt content is low, Na is more soluble and appears in higher concentration than Ca and Mg. When the soil has higher water content and/or is at lower pH, the three salts are in the more soluble forms, which are strongly and positively corredlated to each other. ECe of the soil is controlled predominately by the linear combination of [Na], [Ca] and [Mg]. Of the three salts, Ca has the most effect on ECe. Because ECe is related to the sum of salt concentrations, the expression of salts in the form of SAR_p is not as strongly related to ECe. Only the above 3 highly correlated relationships are recommended.