Centrifuge model tests on slopes subject to shaking and rainfall have been performed to examine the response of slopes with shaking-induced cracks to subsequent rainfall and evaluate the corresponding landslide-triggering mechanisms. The failure pattern of the slope subject to shaking and then rainfall was found different from that of the slope subject to only rainfall. When shaking caused cracks on the slope shoulder and rupture line below, the mobilized soil slid along the slip surface that extended to the rupture line, the main crack became the crown of the undisturbed ground once the slope was subject to a subsequent rain event, and the progression of the landslide was related to the rainfall intensity. During the landslide caused by light rainfall, the main scarp kept exposing itself in the vertically downward direction while the ground behind the main crack in the crack-containing slope remained undisturbed. The detrimental effect of cracks on soil displacement was more evident when the slope was exposed to heavy post-shaking rainfall, resulting in a rapid and massive landslide. Additionally, the volume of displaced material of the landslide, the main scarp area on the upper edge, and the zone of accumulation were larger in the crack-containing slope subject to heavy rainfall, in comparison with those in the crack-free slope. The deformation pattern of slopes with shaking-induced cracks during rainfall was closely related to rainfall intensity and the factor of safety provided a preliminary estimation of slope stability during rainfall. Moreover, even when subjected to the same rainfall, the slopes with antecedent shaking-induced cracks displayed different levels of deformation. The slope that experienced larger shaking had greater deformation under the following rainfall, and the shaking-induced slope deformation also controlled the slip surface location. Finally, the velocity of rainfall-induced landslide could be greatly influenced by the prior shaking event alone. Despite being under light rainfall, the slope that has encountered intense previous shaking exhibited an instant landslide.

Design of retaining walls in clay is typically based on ultimate limit state calculations to prevent collapse with arbitrary factors of safety used to limit deformations. These factors of safety do not take into account the different rates of strength mobilisation in the wide variety of clays found worldwide. As there is substantial uncertainty in this approach, conventional design tends to lead to excessive conservatism with associated high cost. The novel analysis procedure based on the fraction of the strength of soil mobilised for a given wall displacement developed here allows rapid assessment of wall deformations and stresses via a simple two-parameter constitutive model which can be easily calibrated using conventional triaxial data. The model is validated based on field and model case histories with a variety of different clays and propping conditions and is shown to exhibit good performance in predicting the behaviour of published case histories based on soil parameters extracted from previously published soil test data. This novel analysis provides for the first time a route for practising engineers to carry out fast, efficient design at early stages of the design process by considering many potential wall geometries without the computational overhead of complex finite element or finite difference numerical models. KEYWORDS retaining walls; design; clays text Click here to access/download;Main text;A Design Method for Flexible Retaining Walls in Clay_v9.docx

No abstract

No abstract

Stability of slopes where water was accumulated at the toe area during dynamic events were studied in this paper. The soil below and right above water level tended to be contractile and loose due to water storage at the toe of slope, and the slope behaviors became different compared with those without water storage. Several model tests in a geotechnical centrifuge were carried out in this paper to examine the effects of water accumulation at slope toe, relative density of soil, and shaking intensity of dynamic events on the slope stability. The compression of the soil below the water level and localized failure at the toe region subjected slopes to more shearing during shaking and potentially large movements. For the unsaturated slopes with high suction, total collapse was not possible even in model where intense shaking was excited. Successive shaking with increasing intensities could cause more cracks on the surface of the slope, and the deep and wide cracks were mainly on the upper middle part of the slope. Cracks concentrated at areas near the water level. Seepage and capillary flow are two factors that need further attention in the study of the dynamic response of unsaturated slopes with water storage at toe.