An Investigation On The Application Of Two And Three Dimensional Limit Equilibrium Analyses In Slope Stability

Abstract

In this study, it was aimed at comparing the two and three-dimensional limit equilibrium methods, which were applied to different slope geometries and a real case study. Also, each method individually investigated with the aid of computer codes. Four very well-known and one inspired hypothetical slope geometries have been chosen from the literature for the investigations. Many combinations of scenarios have been integrated to these slope geometries including different conditions with groundwater levels, ponded waters, pseudo-static condition and external loadings. Total of 164 scenarios has been modelled in both two and three-dimensions. Morgenstern & Price’s procedure was taken as the reference method as it is one of the well-known methods for its accuracy under various conditions. During the two-dimensional analyses, underestimation and overestimation of the safety factors were noticed in different scenarios with various methods by up to - % 40 and + % 60. The same comparison between the three-dimensional analyses indicated that the safety factors were varying between - % 34 and + % 14 in all of the scenarios. The most intriguing results have been found in between the comparison of two and three-dimensional analyses. The calculated factors of safety were as low as 70 per cent and as much as 540 per cent different in some scenarios. In addition to the hypothetical cases, a landslided area was chosen for the investigation using the limit equilibrium method at Denizkonak region in Kastamonu, Turkey. The investigations were started with a field study following the laboratory tests. The investigated area was generally formed by highly disintegrated flysch with layered siltstones. Therefore, Hoek-Brown failure criterion and Geological Strength Index were chosen for the strength and visual definition of the flysch. Two separate, but interdependent, landslides were observed at the investigated area, and possible cross-sections before the failures were drawn based on the current topography. Back-analyses were performed to obtain the conditions of failure and found that a high pore-pressure ratio was needed to trigger the landslide. However, the second landslide has failed with a lower pore-water pressure and the current topography was determined to be on the edge of failure with a slight increase in the pore-pressure ratio. Also, three-dimensional analyses of the landslide area have generated unconservative results.