Discussion: Determination of Atterberg limits using the vane shear test method [Bull. Min. Res. Exp. (2024) 174: 1-10]

Abstract

This article presents a discussion of the original research reported in the paper by Kayabalı et al. (2024) (the Authors) that was recently published in the Bulletin of the Mineral Research and Exploration Vol. 174, pp. 1–10. Using liquid limit (LL), plastic limit (PL) and vane shear strength against water content [i.e., su(VST) –w] results obtained for 100 fine-grained soil samples, the Authors performed multiple regression analyses to produce four strength-based correlations for predicting the LL and PL values based on su(VST) –w measurements. The Authors’ dataset primarily consisted of residual soils formed through the weathering of igneous rocks, along with a few lacustrine soil samples, all sourced from the vicinity of Ankara, Türkiye. This discussion article examines the veracity of the Authors’ claims regarding the predictive performance of their proposed correlations when applied to other fine-grained soils. Using a sizable independent database of dissimilar fine-grained soils compiled from the research literature, it is conclusively shown that, contrary to the Authors’ assertions, their proposed correlations generally produce poor LL and PL predictions when employed beyond the calibration soil types. This outcome is not unexpected, since the Authors’ data-driven correlations were deduced based on a specific dataset with limited diversity in terms of soil physico-chemical and mineralogical attributes. The article closes with a discussion of the plausible explanations for the poor applicability of the Authors’ correlations when applied to dissimilar fine-grained soils