Day 4

Detailed paper information

Back to list

Paper title Insight into the April 2019 Hoseynabad-e Kalpush landslide disaster in Iran: Results from multi-sensor satellite and in-situ measurements
  1. Magdalena Vassileva GFZ German Research Center for Geosciences Speaker
  2. Mahdi Motagh GFZ German Research Center for Geosciences
  3. Sigrid Roessner
  4. Bahman Akbari Forests, range and watershed management organization
Form of presentation Poster
  • D1. Managing Risks
    • D1.01 Satellite EO for Geohazard Risks
Abstract text The combined effects of extreme rainfall events and anthropogenic activities are increasing the landslide hazard worldwide. Predicting in advance when and where a landslide will occur is an ongoing scientific challenge, which is related to an accurate in time and space analysis of the landslide cycle and a thorough understanding of all associated triggering factors. Between mid-March and the beginning of April 2019, almost the whole of Iran was affected by intense record rainfall leading to thousands of slope failures. In particular, a catastrophic landslide occurred in Hoseynabad-e Kalpush village, Semnan, Iran, where more than 300 houses were damaged, of which 160 completely destroyed. Several questions were raised in the aftermath of the disaster as to whether the landslide was triggered by the heavy precipitation only or by the additional load and seepage from the nearby dam built-in 2013 on the opposite side of the slope.
In this study, we use a multi-scale and multi-sensor data integration approach using satellite and in-situ observations to investigate the pre, co, and post-failure of the Hoseynabad-e Kalpush landslide and assess the role of the potential external factors in triggering the disaster event. Multi-temporal SAR Interferometry observations detected precursory deformations on the lower part of the slope that started in April 2015, accelerated in January 2019 following the exceptional rainy season, and culminated in a slope failure, measured with optical cross-correlation technique, of more than 35 m in the upper part. Subsequently, the lower and middle sections of the landslide showed instability with a maximum cumulative displacement of 10 cm in the first 6 months. To evaluate the role of meteorological and anthropogenic conditions in promoting the slope instability, we integrate the geodetic observations with 20 years rainfall dataset from the Climate Hazards Group InfraRed Precipitation (CHIRP) with Station data, daily in-situ records of the dam reservoir water levels available from September 2014 until August 2019, and cloud-free Landsat-8 images acquired starting from April 2013 integrated with Shuttle Radar Topography Mission elevation data to indirectly estimate the previous to the recorded dam water levels.
The observed pre-failure displacements are a clear indication of the gradual weakening of the shear strength along a pre-existing shear surface or a ductile deformation within a shear zone, which led to the failure. The initialization of the creep followed the reservoir refilling cycle of 2015, while, apart from the final acceleration phase, no clear correlation with the precipitation was observed. The hydraulic gradient due to the dam water level generated a water flow through the porous soil, with field evidence of leakage and piping processes, which permanently altered the hydraulic conditions and therefore mechanical properties of the terrain. Under these already aggravated hydraulic conditions, cumulative rainfall acted from one side by increasing further the reservoir water level, and therefore gradient, and from the other excess pore water pressure on the slope and acting as an additional down driving weight.
While the location of deep-seated landslides can be predicted using only remote sensing geodetic measurements, the time predictivity of the failure is still unreliable especially for slopes where more external factors interact. Hoseynabad-e Kalpush landslide case study has an important relevance also for other parts of the world where artificial reservoirs might act as triggering factors for the slope instability.