The objective of this study is to analyse the relationship between uniaxial compressive strength (UCS) with physical properties, such as apparent density (AD) and ultrasound velocity (PWV), and with other mechanical properties, such as point load test (PLT) and indirect tensile strength (BTS). This analysis was carried out on a group of ignimbrites, consisting of a total of 45 samples, from the Canary Islands and consisted of applying simple regression and multiple regression methods. Applying the simple regression method, it was determined that the ultrasound velocity (PWV) is the property that explains the variability of the compressive strength (UCS) to the greatest extent. Using the multiple regression method, a model capable of increasing the proportion of variability of the (UCS) is obtained, using the apparent density (AD), the ultrasound velocity (PWV) and the point load test (PLT) as independent variables.
Although all properties are important and necessary for a correct physical-mechanical characterisation of the rock mass, the uniaxial compressive strength (UCS) is one of the most relevant properties when estimating its load bearing capacity. This parameter is obtained in the laboratory by the unconfined compression test performed according to the procedure described in the corresponding standard. Usually, is one of the most expensive test on rocks, due to the characteristics of the equipment needed and the time required to prepare the samples, as well as being a destructive technique that makes it impossible to reuse the specimen after testing. This has generated a lot of research, in which the authors propose different indirect methods for estimating UCS and whose predictive functions are derived from simpler tests.
Most commonly, in this type of research, several properties are included in order to determine the best relationship for predicting UCS among all of them. In other studies, in addition to the simple regression method the multivariate regression method is applied in order to estimate UCS using two or more properties as independent variables. The interest in this topic is well demonstrated in the work of Aladejare et al. (2021) a magnificent database which compiles the information available in the literature on the estimation of UCS. The review covers more than 250 references in which authors apply simple and/or multiple regression methods. However, it should be noted that the vast majority of this kind of studies have been carried out on sedimentary or metamorphic rocks, with very few literature on igneous rocks, especially on those of volcanic origin.
