Comparison of site-specific and empirical correlations for drained residual shear strength

Abstract

The discusser has an interest in the topic of the paper (Hayden et al., 2018) on the ability of empirical correlations to predict the residual friction angle of samples from a specific site. In fact, he provided one of the few data sets used by the authors in their study. He also agrees with them that correlations should not entirely replace site-specific laboratory measurement wherever it is available. However, the discusser has a concern on the paper's general conclusion regarding the relatively poor performance of the considered correlations in predicting the measured residual friction angles. In the discusser's opinion, this is in particular not accurate for the correlation by Stark & Eid (1994) and its updated versions (e.g. Eid, 1996; Stark & Hussain, 2013; Eid et al., 2016). As the principal developer of this type of residual shear strength correlation, the discusser has been emphasising the fact that heavily overconsolidated clays, claystones and shales possess diagenetic bonding that results in particle aggregation. This aggregation usually survives the standard sample preparation procedure and consequently influences the measured index parameters. Eid (2001) clearly showed that sand-size and silt-size bits of indurated materials can survive the standard sample preparations for Atterberg limits and hydrometer tests, respectively, and are wrongly considered as sand and silt contents. As a result, these materials should be disaggregated by ball milling until all particles pass the standard sieve no. 200. In most cases, using a mortar and pestle to pass sieve no. 200 is not enough to disaggregate these indurated bits. Extensive experimental studies have been conducted specifically to correlate the indices of the indurated materials to those yielded using the standard procedures (e.g. Eid, 2006). Using the indices derived from the standard procedures influences the accuracy of their correlations to the results of shear strength testing in which the aggregated particles would be battered. Because of the significant effects of this issue on predicting the shear strength friction angles, Eid et al. (2016) and Eid & Rabie (2017) encompassed the equations needed to adjust the index parameters of indurated materials in the figures of their correlations, as shown in Fig. 8 for the residual shear strength.