PM4SAND and UBCSAND Constitutive Soil Models for Seismic Soil-Structure Interaction Analysis of an Anchored Sheet Pile Wall System Viet Tran, Ph.D, P.Eng Stantec Consulting Ltd. 500-4730 Kingsway, Burnaby BC V5H 0C6 Canada Email: viet.tran@stantec.com ABSTRACT Seismic soil-structure interaction analyses were performed to evaluate seismic performance of anchored sheet pile wall system. The analyses were carried out using UBCSAND and PM4SAND soil constitutive models for liquefiable soils. Model calibration was performed to determine input parameters for the seismic analysis. The calibration considered some important aspects that significantly affect the results of the numerical simulation, including cyclic resistance ratio, the development of the excess pore water pressure, the overburden effect, the static shear stress effect, and the modulus reduction and damping. Soil liquefaction and seismic response of the sheet pile wall system estimated using PM4SAND model agree reasonably with the those estimated using UBCSAND model. Use of different soil models with proper model calibration can capture possible range of the expected response. RÉSUMÉ Des analyses d'interaction sismique sol-structure ont été réalisées pour évaluer la performance sismique du système de mur de palplanches ancrées. Les analyses ont été effectuées à l'aide des modèles constitutifs des sols UBCSAND et PM4SAND pour les sols liquéfiables. L'étalonnage du modèle a été effectué pour déterminer les paramètres d'entrée pour l'analyse sismique. L'étalonnage a pris en compte certains aspects importants affectant de manière significative les résultats de la simulation numérique, notamment le rapport de résistance cyclique, l'augmentation de la pression interstitielle, l'effet de surcharge, l'effet de cisaillement statique et la réduction et l'amortissement du module. La liquéfaction du sol et les réponses sismiques du système de palplanches estimées à l'aide du modèle PM4SAND concordent raisonnablement avec celles estimées à l'aide du modèle UBCSAND. L'utilisation de différents modèles de sol avec un bon calibrage du modèle peut capturer la portée possible de la réponse attendue. 1 INTRODUCTION Liquefaction of soil presents a significant hazard to waterfront structures. Finite element or finite difference analysis for soil liquefaction during earthquake can be assessed using advanced soil constitutive models which consider the development of excess pore water pressure. UBCSAND (Beaty and Byrne, 2011), PM4SAND (Boulanger and Ziotopoulou, 2017), Pressure Dependent Multi Yield 02 (Yang et al., 2008), and SANISAND (Dafalias and Manzari, 2004; Taiebat and Dafalias, 2008) are several constitutive models that have been developed for liquefaction simulation. The UBCAND and PM4SAND models have been used in engineering practice. UBCSAND is a two-dimensional effective stress plasticity model that predicts the shear stress-strain behavior of the soil using an assumed hyperbolic relationship and the build-up of excess pore water pressure during cyclic loading (Beaty and Byrne, 2011). PM4SAND is a critical state compatible, stress-ratio controlled, bounding surface plasticity model, developed to approximate the range of behaviors important to geotechnical earthquake engineering practice (Boulanger and Ziotopoulou, 2017). Using both models can provide insight into the model response and suggest the approximate range of simulation results. Poor agreement between results obtained using two different models can identify issues with the simulation and lead to subsequent improvements. In this study, seismic soil-structure interaction analyses were performed to evaluate seismic performance of an anchored sheet pile wall system. The soil profiles and structure properties were adapted from published literature and do not represent any local site or structure. Sandy soils were modeled using UBCSAND and PM4SAND soil constitutive models to analyze the development of excess pore water pressure. 2 SUBSURFACE CONDITIONS The soil profiles at the site are shown in Figure 1. The total height of the steel sheet pile wall is about 22.5 m and the top of the wall is at approximately El. + 2.0 m. The sheet pile wall is connected to an anchor wall using high strength steel tie rods which are installed near top of the wall and spaced at 2 m. The anchor wall is located 20 m behind the sheet pile wall. The natural subsurface soils consist of 8 m to 20 m thick dense sands, overlying a 2.5 m thick clay layer which is underlain by dense silty sand to sand. The backfill behind the wall typically consist of sand with thicknesses varying from approximately 12 m right behind the sheet pile wall to 8 m at the anchor wall location. The backfill sand is estimated to be loose with a fines-corrected (N1)60cs value of 9.