A flowslide overriding liquefied substrate can vastly enhance its disaster after failure initiation, due to rapid velocity and long-runout distance during landslides mobilized into flows. It is crucial to provide improved understanding to the mechanism of these catastrophic flowslides for hazard mitigation and risk assessment. This study focuses on the Saleshan landslide of Gansu in China, which is a typically catastrophic flowslide overrode a liquefied sand substrate. Geomorphologic and topographic maps along with analysis of seismic signals confirm its dynamic features and mobilized behaviors. ERT surveying detected abundant groundwater in the landslide, which is fundamental to its rapid long-runout distance. Particle size distributions and triaxial shear behaviors affirmed more readily liquefied behavior of superficial loess and underlying alluvial sand than red soil sandwiched them. We also examined the liquefaction susceptibility of the alluvial sand under loading impact at undrained and drained conditions. The alluvial sand is readily liquefied in the undrained condition while it is difficult at drained condition due to rapid water pore pressure dissipation. The results showed that the landslide experienced a sudden transformation from slide on the steep slope where it originated to flow on a nearly flat terrace with abundant groundwater that it overrode. This transformation can be attributed to the liquefied alluvial sand substrate enhancing the whole landslide body mobility. Along with recent, similar findings from landslides worldwide, substrate liquefaction may present a widespread, significant increase in landslide hazard and consequent mobility and our study reveals conditions necessary for this phenomenon to occur.

The Heifangtai area is commonly known as the museum of loess landslides in China. Irrigation-induced loess flowslides frequently recur along the margin cliffs of the Hefaingtai terrace, causing 42 fatalities and significant economic losses, as well as major ecological and environmental problems, such as increased soil erosion rate. The initiation and mobility of these irrigation-induced loess flowslide recurrences remain undetermined. On three typical recurrences of the loess flowslides, we performed joint geophysical detection using electrical resistivity tomography (ERT) and multichannel analysis of surface waves (MASW), and also tested loess basic properties by field profile sampling. In addition, we examined the shear behaviors of saturated loess utilizing an undrained ring shear apparatus. The geophysical signatures and in-situ loess property profiles showed that hydrogeological conditions are key to the initiation of recurring loess flowslides. The results also demonstrated that liquefaction shear behaviors of saturated loess control the mobility of after-failure of the loess flowslides. Rapid criteria of liquefaction susceptibility evaluation are suggested to provide a better understanding of the dynamic mechanisms of loess flowslides. These findings shed substantial light on long-runout flowslides that occur in fine-grain soil and their implications for landslide hazard mitigation.

Gully Stabilization and Highland Protection (GSHP) techniques are useful in preventing gully erosion and have been widely utilized in the Loess Plateau. Rolling backfill is used to fill ditches in remolded loess, which is an important part of gully stabilization and highland protection, but destroys the original loess structure and changes the circulation of groundwater and surface water leading to a rise in groundwater. Groundwater rising is an important factor for filled loess slope instability and can induce landslides. A test device was designed to study the process of water infiltration into the filled project and the failure process of the filled loess slope due to groundwater rising. First, the groundwater was uniformly infiltrated with water, then preferential seepage with the deformation and cracks appeared in the slope. The pore-water pressure response to the groundwater infiltration and the pore-water pressure in the front of the slope body sharply increased, especially near the sliding surface, while the pore-water pressure at the back of the slope sharply decreased during slope failure. The failure process of the experimental slope can be divided into three stages: settlement deformation, collapse deformation, and slope toe slide-flow or regressive failure. In the first and second stages, the deformation is vertical displacement as slope settlement, and the third stage deformation is mainly horizontal displacement in the direction of the free surface of the slope body. The filled slope failure is due to groundwater infiltration with suffusion erosion, saturated softening, and infiltration dynamics.