Open pit slope design in the bedded units encountered in the iron ore pits of Australia’s Pilbara region is challenging in many aspects. Data collection and interpretation, slope stability assessment, and implementation of design parameters in these anisotropic rock masses require specialised experience, as there are numerous complex considerations. Evaluation of bedding shear strength is a critical aspect in the process, which presents challenges in collecting appropriate samples; successful execution of laboratory testing; appropriate interpretation of the effects of infilling; and interpreting the impact of the undulations and asperities on the bedding surface. Back-analysis of failures along bedding planes and in-pit large-scale shear testing are seldom an option. The selection of an appropriate strength model for each stability analysis section depends on the rock units present and the anisotropy orientation.
2D limit equilibrium methods, incorporating the Snowden modified anisotropic linear model, are commonly used for stability assessments. The outcomes using this method are sensitive to analysis methods, slip surface search limits and methods, and key model inputs. It is important that the possible failure mechanisms are well understood and suitably analysed. A critical part of the design process is the identification of design domains and the selection of representative sections for slope design analyses. SRK has developed refined methods in this regard. In some cases, more advanced stress-strain based numerical programs, such as finite difference or distinct element codes, should be used.
It is a common assumption in anisotropic 2D analyses that slopes will be fully bedding-controlled where bedding strike is ≤30° to the strike of the slope, and that stability will be best represented by isotropic (rock mass controlled) conditions where bedding is >30° in obliqueness. However, 3D analyses can better capture failure mechanism and instability risk in the context of rock mass confinement and bedding obliqueness. SRK has performed 3D analyses that have allowed better understanding and prediction of these controls, for consideration where only 2D analyses are able to be carried out. When the bedding dip angle is greater than its shear angle, the difference between the 2D and 3D factor of safety (FoS) of a slope increases significantly as bedding obliqueness increases. The FoS in 3D is slightly higher than in 2D for bedding that is parallel to the slope. However, where the bedding strike is only 10° oblique to the slope, the difference in FoS is already significant (≥15%) and increases to >20% from 20° upwards. This demonstrates that the results of 2D analyses for bedding obliqueness ≤30° are conservative, and 3D analyses may indicate opportunities for slope steepening. Conversely, the 2D analysis of isotropic conditions where bedding obliqueness is > 30° may significantly overestimate the stability of a design (by >20%), presenting increased risk of failure.
