Investigating the effect of depth on modulus of deformation of rock masses with low strength

Rock mass deformation modulus, which has a very important role in engineering design and which is the parameter that controls the deformation behavior of rock masses, could be obtained with empirical equations obtained with various approaches besides in – situ tests. In this study, the effect of depth and vertical stress on the estimation of deformation modulus was investigated. For this purpose, 12 new prediction models were developed by using different rock types. The database used to develope the prediction models consisted of parameters such as Rock Mass Deformation Modulus, Rock Quality Designation (RQD), Rock Mass Rating (RMR), Uniaxial Compressive Strength (UCS), Intact Elasticity Modulus (Ei) and unit weight, which have been evaluated on 4 different study areas of the İstanbul province. For the descripted parameters, outlier analyses were carried out in order to increase the reliability of the data and in order to determine the extreme values against the general distribution. Using the refined database, deformation modulus values were calculated with the help of the 10 existing empirical equations commonly used in the literature by comparing calculated along with discussing estimated deformation modulus values and the performances of these equations. In order to investigate the relationship between deformation modulus and depth – vertical stress, correlation and regression analyses were carried out. In the correlation analysis, it was found out that the rock mass deformation modulus and RMR given significant correlations with depth – vertical stress. On the other hand, it was found out that the intact rock parameters led to less significant correlations with depth/vertical stress. Within the simple regression analyses, correlation coefficients of 0.40 and above were calculated for most of the variable pairs. With the aim of prediction of the rock mass deformation modulus with depth – vertical stress and other rock parameters, multivariate linear and nonlinear regression models were utilized to establish the final prediction models and hence, to validate the reliability of the empirical equations. For the obtained models, it was determined that the best perfomance for prediction of the deformation modulus was obtained in models used depth – vertical stress and RMR score. For the 12 different empirical equations obtained, correlation coefficients were determined to vary between 0.40 and 0.84. In order to assess the prediction performance of the empirical equations obtained in the present study, VAF and RMSE indices were calculated. According to the RMSE and VAF values, it was found out that the prediction performances of the empirical equations were high and hece, acceptable. For the generated equations, some limitations existed in regards to the utilization of the equations in shallow depths (< 5m) and for massive rock conditions. As a conclusion, it was determined and empirically validated that the variation of rock mass deformation modulus with depth – vertical stress wass evident.