Variability of Soil Characterization Analyses and Impacts on Soil Classification

Abstract

Increased reliance on laboratory-determined soil properties for soil classification has focused attention on the variability of analytical results among laboratories and the potential impact on placement of soils in classes of the U.S. system of soil taxonomy. Eleven laboratories analyzed 20 samples selected from pedons in the USA and Guam to examine the variability of analytical results when the laboratories use standard procedures for soil characterization. Analyses included in this study were extractable cations, cation exchange capacity (CEC) by three methods (NH4OAC, NaOAc, and sum of cations), particle size distribution and pH. The coefficients of variation (CV) provide a measure of the variability of the results. The CV varies for each soil for each analysis. The determination of BaCl2-TEA (pH 8.2) extractable acidity produced the highest CV (up to 190%); determination of pH produced the lowest CV (≤7%). The determination of CEC by NH4OAc and NaOAc produced similar CV (up to 47%). The measurement of clay was the least variable of the particle sizes (CV up to 84%); silt was the most variable (CV up to 141%). These variations in analytical results could result in the placement of the same soil in more than one taxon. The consistent differentiation of Ultisols from Alfisols may be most difficult, given the large variation in the determination of extractable acidity and its effect on the calculation of percent base saturation (CV up to 172% for sum of cations). The large variability among laboratories may reduce the reliability of soil characterization databases compiled from several laboratories.