Modelling of vertical piles subjected to a clay crust movement over a liquefied soil Ripon Karmaker & Bipul Hawlader Memorial University of Newfoundland, St. John’s, NL, Canada Sujan Dutta Terraprobe, Brampton, ON, Canada ABSTRACT Piles might be subjected to passive loading from permanent ground deformations. The present study investigates the lateral force on a pile resulting from a downslope displacement of a clay crust over a liquefied soil layer. Finite-element (FE) analyses are performed to calculate the total force and its variation with depth in the segment of the pile in the crust. Analyses are performed for a single row of piles with a varying centre-to-centre spacing and undrained shear strength of clay. The FE results show that the pile behaves as a single pile when the spacing is greater than five times its diameter. The arching effects in relation to pile spacing are discussed. The lateral force per pile decreases with a decrease in pile spacing. Using the calculated maximum lateral force, a separate analysis is performed to examine the structural response of a long pile installed through the liquefied layer on a stable soil layer. RÉSUMÉ Les piles pourraient être soumises à une charge passive due à des déformations permanentes du sol. La présente étude étudie la force latérale sur une pile résultant d'un déplacement en pente d'une croûte d'argile sur une couche de sol liquéfié. Des analyses par éléments finis (EF) sont effectuées pour calculer la force totale et sa variation avec la profondeur dans le segment de la pile dans la croûte. Les analyses sont effectuées pour une seule rangée de piles avec un espacement centre-centre et une résistance au cisaillement non drainé variable de l'argile. Les résultats FE montrent que la pile se comporte comme une seule pile lorsque l'espacement est supérieur à cinq fois son diamètre. Les effets d'arche en relation avec l'espacement des poils sont discutés. La force latérale par pile diminue avec une diminution de l'espacement des poils. En utilisant la force latérale maximale calculée, une analyse séparée est effectuée pour examiner la réponse structurale d'un long pieu installé à travers la couche liquéfiée sur une couche de sol stable. 1 INTRODUCTION Piles can be subjected to two different types of lateral loads. In active pile loadings, the lateral forces, which might come from superstructures, create a load on the pile and then transfer to the surrounding soil through pile–soil interaction. In this case, the soil surrounding the pile provides a resistance to the movement of the pile. In passive piles, the displacement of a layer/block of soil near the ground surface creates a load on the pile, which is then transferred to the deeper soil layers through pile–soil interaction. The ground deformation could be caused by slope failure, lateral spreading due to the formation of a weak failure plane or liquefaction of loose sand layer(s) due to an earthquake. In many cases, a non-liquefied soil layer above the liquefied sand layer/weak zone displaces a significantly large distance, especially in a slopping ground condition, even for a mild slope. For example, Cubrinovski et al. (2009) reported permanent lateral ground displacements of up to 4 m in some mild-sloped areas after the 1995 Kobe earthquake. The displacement of soil caused significant damage to piles in those areas. The upper non-liquefied crust could be cohesionless, cohesive, or c- soil. The present study focuses on the estimation of lateral force that could be exerted on a pile due to downslope movement of a clay crust. As this type of ground deformation occurs quickly (e.g., during an earthquake or post-quake deformation), the soil is modelled for an undrained condition. Physical and numerical modelling have been performed in the past to understand the response of piles in clay under active lateral loadings. For example, Welch and Reese (1972) and Matlock (1970) presented the response of instrumented piles under lateral loadings. Based on field test results, lateral load per unit length (p) versus displacement (y) curves have been developed to calculate the structural response of the pile. Centrifuge tests were also conducted to model the lateral pile–soil interaction (e.g., McVay et al. 1998 and Taghavi et al. 2016). Moreover, analytical solutions have been developed to calculate the maximum lateral resistance of clay on a section of the pile. For example, Randolph and Houlsby (1984) developed a closed-form solution for a single pile where the clay was modelled as an isotropic rigid-plastic Tresca material. The ultimate resistance has been presented in a normalized form as N = P/suD, where P is the lateral capacity per metre length of the pile, D is the diameter of the pile and su is the undrained shear strength of the clay. It has been shown that N = 9.14 and N = 11.94 for the fully smooth and perfectly rough pile–soil interface condition, respectively. Empirical formulas, analytical solutions, and numerical techniques have been developed to calculate the passive load on vertical piles in clay. Conducting small-scale physical model tests, Bauer et al. (2014) showed a wide variation in the lateral force when a kaolin clay block interacts with a single pile or rows of piles. A summary of