The present research study aims to create accurate and comprehensive inventory mapping while investigating the geomorphological and geotechnical characteristics of the large, deep-seated, and damaging El Kherba landslide triggered by the August 7, 2020 (Mw 4.9) Mila earthquake. The methodology relies on the analysis of results obtained through detailed field investigations, satellite image interpretation, deep boreholes equipped with piezometers, laboratory tests, in situ tests, and numerical simulations. The resulting landslide inventory map reveals a significant earth slide with an active zone covering a surface area of 1.565 km2, extending approximately 2,166 km in length, with a width ranging from 40 m to 1.80 km, and a volume of 25,784,909 m3. Geomorphological field mapping results revealed a large and deep-seated morphological deformation related to: (i) the weak mechanical resistance and low stability slopes that the seismic strengths caused a reduction in the shear strength of the soil; (ii) Miocene clays, highly altered and potentially subject to shrinkage and swelling; (iii) a partial reactivation of a previously existing large landslide; (iv) human activity such as slope excavation and unplanned urbanization; and (v) topographical and lithological site effects. The results of geological and hydrogeological investigations indicated the presence of: (i) thin and thick weak-resistance interlayers of altered and plastic clays with weak resistance, which may constitute shear surfaces; (ii) a shallow aquifer that impacted the mechanical resistance characteristics. Laboratory tests revealed that the fine clay in the soil was highly weathered, with a low dry density and a high moisture content, along with a high saturation and plasticity, making it very sensitive to the presence of water. Undrained triaxial cyclic loading tests indicated a high potential for the generation of excess pore-water pressures in the material during seismic loading. The direct shear test showed that the disturbed soils had an average cohesion of 33.4 kPa/m2 and an internal friction angle of 18.21 degrees, indicating poor structural and shearing strength. The results of the oedometer test indicated that the soils are compressible to highly compressible, overconsolidated, and have the potential for swelling. According to the Manard pressuremeter test (MPT) and available empirical relationships, the landslide exhibited a deep-seated nature, with sliding surfaces located along weak geotechnical characteristics interlayers at a depth ranging between 10 and 40 m. The depths of failure obtained from the MPT were consistent with those determined by the empirical relationships available in the literature and numerical simulations. This comprehensive research provides valuable data on earthquake-induced landslide and can serve as a guide for the prevention and mitigation of landslide risks.

The hydrological response of groundwater to rainfall plays a key role in the initiation of deep-seated bedrock landslides; however, the mechanisms require further investigation due to the complexity of groundwater movement in fissured bedrock. In this study, an active translational landslide along nearly horizontal rock strata was investigated. The hydrological response of groundwater to rainfall was analyzed, using the data from a four-year real-time field monitoring program from June 2013 to December 2016. The monitoring system was installed along a longitudinal of the landslide with severe deformation and consisted of two rainfall gauges, nine piezometers, three water-level gauges, and two GPS data loggers. Much research effort has been directed to exploring the relationship between rainfall and groundwater response. It is found that both the pore-water pressure (PWP) and groundwater level (GWL) responses were significantly influenced by the rainfall pattern and the hydrological properties of the underlying aquifer. The rapid rise and fall of PWP and GWL were observed in the rainy season of 2013 with high-frequency, long-duration, and high-intensity rainfall patterns, especially in the lower of the landslide dominated by the porous aquifer system. In contrast, a slower and prolonged response of PWP and GWL to rainfall was observed in most monitoring boreholes in 2014 and 2015 with two rainstorms of short duration and high intensity. In the lower of the landslide, the peak GWL exhibited a stronger correlation with the cumulative rainfall than the daily rainfall in a single rainfall event whereas the peak groundwater level fluctuation (GWLF) exhibited a strong correlation with API with a half-life of 7 days. In the middle of the landslide, however, relatively lower correlation between rainfall and groundwater response was observed. Three types of groundwater flow were identified based on the recession coefficients of different segments of water-level hydrographs in the landslide area, corresponding to the quick flow through highly permeable gravely soil and well-developed vertical joints in the bedrock, the slow and diffuse flow through the relatively less-permeable bedrock, and the transition between them in the aquifer system.