Reply to Discussion of “A Validation Study for the Estimation of Uniaxial Compressive Strength Based on Index Tests” by Kong and Shang, Rock Mech and Rock Eng (2018)

Abstract

In our technical note, we adopted two index tests (point load test and Schmidt hammer test) to estimate the uniaxial compressive strength (UCS) using “standard” and apparently homogeneous brick specimens. The main conclusion reached was that a “perfect estimate” of UCS is possible if a homogeneous rock material is used when performing the index tests, although in many cases laboratory rock samples contain many micro-fractures or pores which significantly affect the performance of the index tests. In the “Discussion”, two main issues are raised. First, many references are given to justify when and why the index tests should be used and it is pointed out that cylindrical cores can be easily drilled from the bricks used in the paper as well as from similar uniform and homogeneous rocks (Yilmaz and Yucel 2014). It is also argued that in these cases there is no need to use the index tests to estimate UCS and that the high correlation coefficient (R2 = 0.99) obtained questions the applicability of the proposed equation. The authors would like to acknowledge Dr. Yilmaz’s comments. The authors do agree with the first comment. The index tests should be used in cases where standard cores are extremely difficult to be prepared in the laboratory, due to the presence of massive fractures or because of the extremely hard nature of some rocks such as cryptocrystalline flint, with an average UCS of ~ 600 MPa (Aliyu et al. 2019). It is also worth to note that the Schmidt hammer tests allow a quick estimate of rock strength for preliminary evaluation purposes in the field, which probably is one of the main reasons for its current popularity. In this reply, the authors would also like to highlight the main objective of the technical note, which for the first time presents some new insights into the validity of the index tests for the estimate of UCS. As shown in the paper, a series of bricks were used to provide a uniform fine-grained and homogeneous material, thus removing the effect of lithological heterogeneity. Considering the scope of the technical note, the index tests were performed on some uniform and homogeneous bricks, although “standard” cores could be easily drilled. It is worth to mention that direct UCS tests were also performed on some cylindrical samples (see Fig. 2 in the technical note) for validation purposes. The reason of using bricks is that the results obtained may vary significantly due to lithological heterogeneity arising from geological bedding and schistosity, grain size variation and micro-fractures. As such, the reliability of the previously derived empirical equations between UCS and index test results are questionable, due to the intrinsic limitation (if any) of the index tests used. Although the relationship between UCS and the results of the index tests has been widely discussed (Hoek 1977; Aggistalis et al. 1996; Fener et al. 2005; Karaman and Kesimal 2015; Aliyu et al. 2019), the validity of the index tests, however, remains poorly understood and this motivated the study carried out. It is as well clear why the correlation coefficient (R2) in the Eq. (2) in the paper can be 0.99, which again attracted the attention of Dr. Yilmaz. The data were based on homogeneous bricks without considering the effect of lithological heterogeneity. The correlation equation, therefore, reflects the “real performance” of the point load test, which has tested the authors’ hypothesis. Although this point might be anticipated to some degree, it still lacks a real justification, as mentioned in the “Discussion” that no related literature showing the strong correlation coefficient (i.e., ~ 1) is available. * Junlong Shang [email protected]; [email protected]