Hide/Show Apps

Comparison of limit equilibrium and 2-d, 3-d finite element slope stability models: a case study on the slope in Akpınar district, in Ankara

Etiz, Mehmet Can
This study presents comparisons of Two-Dimensional (2D) Limit Equilibrium Method (LEM), Finite Element Method (FEM), and Three-Dimensional (3D) FEM slope stability models for the slope in Akpınar District in Ankara. A landslide has occurred on the aforementioned slope following heavy rain in June 2011. One of the buildings on the slope has overturned about 3 to 5 degrees and some other buildings in the area have suffered substantial damages. Following the landslide, several field investigations and subsequent studies have been conducted by geotechnical experts and academicians. In this research, previous studies have been summarized, velocity-time plots have been constructed from the site inclinometer data and the performance of the slope has been studied by different methods on a comparative basis as a case. In the first part of the study, namely static analyses, back-analyses have been conducted to obtain equilibrium strength parameters. Also, the possible effects of the foundation excavation which existed on the toe of the landslide have been investigated. The impact of heavy rain on the slope has been modeled with 2D-LEM and 2D-FEM while the equilibrium condition has been modeled with 3D-FEM. For the seismic stability assessment of the slope, the effects of possible earthquake scenarios have been modeled. Both short term and long-term parameters have been used in the analyses. Pseudo-static coefficients for 2D-LEM, pseudo-static coefficients, and time-history analysis methods for 2D-FEM have been utilized for dynamic calculations. Consequently, it has been shown that deep and shallow sliding surfaces were triggered on the slope as a result of the heavy rain and that the existence of foundation excavation had an unfavorable effect on the initiation of the deep slide surface. Besides, almost in every case for the slope, FEM has provided approximately 10% lower safety factors compared to those offered by LEM. Also, for pseudo-static approach short term parameters and for time-history analyses, long term parameters have provided safer results.