REGIONAL PROBABILISTIC LANDSLIDE HAZARD ASSESSMENT FOR THE ENGURI DAM (JVARI, GEORGIA)

The Enguri Dam forms a reservoir in a seismically active area in the foothills of the Caucasus Mountain range near Jvari, Georgia. The slopes are steep, highly fractured, and weathered, which make them at risk to failure during or following extreme rainfall events. Hydroelectricity produced by the water retained by the 271-m dam provides almost half of the electricity for the country. The importance of hydropower generation is an important reason that this project is supported by the NATO Science for Peace and Security Programme and involves scientists from 7 countries. The reservoir perimeter is more than 40 km and the surrounding slopes span an area of more than 30 square kilometers. The size of the area and paucity of slope data have made slope-failure hazard assessment of the broader area impossible and to date only limited work has been completed and it was focused on a single creeping landslide. Our project evaluated the landslide hazards for the reservoir area using data from past studies, field investigation, and remotely sensed inputs, integrated with Geographic Information System (GIS) based slope stability analysis. The GIS-based Probabilistic Infinite Slope Analysis modeling (PISA-m) program was used to evaluate both the static and seismic slope stability of the region. The geotechnical properties (e.g., unit weight, the angle of internal friction, cohesive strength, and moisture content) were obtained from past literature and field data collection. The remotely sensed Normalized Difference Vegetation Index (NDVI) derived from Landsat 8 was used to account for the vegetation distributions in calculations of root strengths for the slopes. The uncertainties in the input parameters were estimated using extreme value distributions. The static and seismic slope stability analysis reveal that the areas proximal to the dam have low safety factors against sliding and are very susceptible to slope instability, especially to seismic events. The verification of the modeled stability with the landslides mapped using high-resolution remotely sensed data and fieldwork indicates that the PISA-m provides a promising program for regional slope stability analysis.