Relationships between base saturation, effective base saturation and soil pH as the references for the development and verification of criteria for international soil classification

Abstract

Base saturation (BS) and effective base saturation (BSe) are widely used soil characteristics, commonly involved in the criteria of soil classification, including the World Reference Base for Soil Resources (WRB) system. However, their parallel use can result in an internal inconsistency in assessing the base status in particular soil types. Furthermore, numerous methods of estimating BS and BSe lead to an incomparability of analytical results and improper soil classification; therefore, replacement of Bs and BSe with simple pH measurement is postulated. The aim of the present study was (i) to analyse the relationship between BS and BSe and the consequences of replacing BS with BSe in selected classification criteria of the WRB system, and (ii) to analyse the relationships between pH and BS and BSe, and the possibility of replacing these measures with pH for at least provisional soil classification. An analysis was carried out using a large database (more than 290,000 soil horizons) compiled from the datasets from the USA, Canada, Portugal, Poland, ISRIC, and other published papers, which comprises soil representing various climate zones, parent materials, and soil types. In mineral soils, 50% BS corresponds to 75–77% (in Andosols: 85%) BSe, therefore, BS and BSe cannot be equivalently replaced in the requirements for Dystric and Eutric qualifiers. Setting the criteria at 50% BSe in the latter editions of WRB results in a possible overestimation of the abundance of soils with Eutric qualifier compared to estimations based on the criteria of the FAO Legend to World Soil Map and early editions of WRB. The study confirmed the statistically significant relationship between BS and pH and estimated the pH_w values corresponding to 50% BS and 50% BSe at 5.2 and 4.7, respectively, in mineral soils. The large variation of these relationships justifies separate thresholds for organic soils and Andosols (at pH_w 4.9 and 6.1, respectively) and the differentiation of pH thresholds for the remaining mineral soil in relation to soil organic carbon and clay content. The pH_KCl values corresponding to 50% BS and 50% BSe were estimated with lower determination coefficients than for pH_w. Thus, setting the thresholds on pH_KCl values seems less reliable and more difficult as an in-field procedure compared to pH_w.