Characterizing the spatial distribution and fundamental controls of landslides in the three gorges reservoir area, China

Abstract

A comprehensive field investigation and analysis of the spatial distribution of landslides was conducted in the Three Gorges Reservoir Area (TGRA) along the Yangtze River and its first-order tributaries. A landslide inventory for this area was prepared using regional-local geological investigations and other available information. The analysis of the extensive inventory data reveals that the occurrence of landslides is mainly affected by four primary factors: (1) Lithology—the highest incidence of landslides is associated with the lithological combination of marl and shale intercalated with mudstone (MSM), and a number of landslides occurred in an area dominated by sandstone and mudstone intercalated with shale and coal seams (SMSC), which cover a major part (approximately 71%) of the bank slopes; (2) Structure—slope structure fundamentally controls the development of the sliding zone and the mechanism of formation of landslides occurring on the same lithological combinations at subregional scales. Landslide concentrations in consequent slopes are approximately 5–17 times greater than those of reverse slopes and diagonal slopes; (3) Topography—landslides are particularly abundant between elevations of 100 m and 600 m. The elevations of the headscarps of the landslides gradually decrease from the head to the tail of the reservoir due to the topography, and almost all of the toes of the landslides are below the highest historical floodwater level. The analysis of the correlation between landslide occurrence and hillslope gradients shows that 50% of all landslides occurred on slopes between 10° and 15°, despite slopes in this range representing only 30% of the total hillslope gradients; and (4) Water level fluctuation and precipitation—the fluctuation of reservoir water level exerts a fundamental effect that controls the occurrence and reactivation of landslides during the period of initial impoundment, whereas the precipitation becomes even more important than water level fluctuation after the first impoundment at 175 m above sea level. In addition, the comprehensive analysis performed in this study provides broad insight into the landslide distribution in the TGRA. This work will improve our understanding of the role of morphological and geological conditions that affect the occurrence of landslides, which may provide useful guidance for regional risk assessment and landslide susceptibility studies.