Relationship between 1:5 Soil/Water and Saturated Paste Extract Sodium Adsorption Ratios by Three Extraction Methods

Abstract

Cations extracted from soil using non-standard techniques are used to calculate the sodium adsorption ratio (SAR). To interpret these values, analytical approaches for converting alternative approaches to standard approaches are needed. The objective of this research was to develop the relationship between the standard approach (saturated paste extract, SARe) and alternative approaches where the cations are in 1: 5 soil/water ratios and are mixed by shaking, stirring, or an NRCS method (allowed to reach equilibrium). One hundred soils sampled from glacial parent materials in North Dakota were selected for this study. The SAR values from the three methods were highly correlated to each other. Simple linear regression (Model 1), robust regression, and Model 2 were analyzed for the relationship between SAR1:5 and SARe. Outlier analysis and calcite content distribution indicated the possible influence of calcite content. The soil data were classified into high and low calcite by a 4.2% criterion. In addition, Ca in the 1:5 and saturated paste extracts showed poor relationships, also indicating that Ca had an influence. Model 1 and robust regression were similar in expressing the relationship between soil SAR1:5 and SARe with a normal residual distribution, but Model 2 had high left skewness in the residual distribution. The model prediction was increased when soil data were classified by calcite content. Predicting the SARe of soils from SAR1:5 data is possible but soil-calcite concentrations should be considered; 1:5 methods can confidently be compared.