engleski

Slope stability analyses of the Kostanjek Landslide for extreme rainfalls in the winter of 2013

The Kostanjek landslide in the City of Zagreb is one of the pilot areas within the scientific Croatian-Japanese SATREPS FY2008 project ‘Risk Identification and Land‐Use Planning for Disaster Mitigation of Landslides and Floods in Croatia’. It is located in the western part of the City of Zagreb, in residential area at the base of the south-western slope of the Medvednica Mt. The Kostanjek landslide was activated in 1963 by excavation in two open pit mines. Massive blasting and excavation in the foot of the slope caused slope movements within an area of 1.2 square kilometres. A landslide model was proposed by Ortolan (1996) on the basis of geotechnical investigations from 1984 to 1989 . According to this model there are three sliding surfaces at different depths. The depth of the deepest sliding surface is about 90 m ; the depth of the intermediate sliding surface is 65 m, while the superficial sliding surface is about 50 m deep. In the frame of the Croatian-Japanese project on the Kostanjek landslide monitoring system was established. The system consists of sensors for measuring displacement, hydrological properties and external triggers (Mihalić et al., 2013). Based on the monitoring data from GNNS sensors, extensometers, borehole extensometers and inclinometers it was concluded that Kostanjek landslide is deep-seated large translational landslide with one sliding surface. This paper presents results of slope stability analyses performed using the LS-RAPID software (Integrated Landslide Simulation Model). Unlike the Ortolan’s model, where the sliding surface reaches maximum depth of the 90 m in the southeastern part of the landslide body, in these analyses the sliding surface was created in the form of ellipsoidal sliding surface with maximum depth at the central part of the landslide body. According to the previous analyses (Gradiški et all. 2013) it was concluded that it is more likely that sliding surface is not so deep in the south-eastern part of the landslide body. For reliable interpretation of the sliding surface depths, additional subsurface investigations are necessary. Parameters used for these analyses were determined from drained test of samples in ring shear apparatus. The samples were taken from the core on the assumed position of the sliding surface at the depth of around 65 m. The testing of the samples was carried out by Maja Oštrić in the frame of the Croatian- Japanese SATREPS FY2008 project. The extreme rainfalls in the winter of 2013 caused a reactivation of the Kostanjek landslide. Monitoring sensors at the Kostanjek landslide recorded greater displacements in the central part of the landslide than the displacement at the landslide borders (Krkač et all, 2013). The meteorological data from the hilly area of the City of Zagreb shows the following extremes (Bernat et all. 2013): cumulative monthly precipitation in January, February and March in 2013 was two to three times higher than the average monthly values for the period from 1862 to 2012 ; and cumulative precipitation for a 3-month period in 2013 has the highest value in the last 150 years. Taking into account the extreme rainfalls in the winter of 2013 it was assumed that the sliding surface was fully saturated and the pore pressure ratio was set as a fluctuating value and it was increased from 0.3 to 0.8. In these analyses only the rainfalls was taken as triggering factor for the landslide occurrence. According to the results of the analyses, the more unstable part of the landslide is the central part of the landslide body, i.e. the slopes of the abandoned old open marl pit. In the analyses the movements started at the central part of the landslide body after that the failure area will expand around the initial failure zone. At the end of simulation the area of the whole landslide mass corresponds to the landslide contour from Ortolan’s model (from 1996) and to the new landslide deformations (cracks, bulging, subsidance) identified in the field after March 2013 (Fig.1). These analyses were performed for the assumed pore pressure on the sliding surface. For more precisely analyses it is necessary to define more preciously: sliding surface with more correct positions along particular landslide cross sections ; water table surface derived on the basis of measured water levels ; and performed undrained ring shear test to determined more appropriate soil parameters.

Kostanjek Landslide; LS-RAPID; extreme rainfall

nije evidentirano