Shear Stress Used in Soil Liquefaction: Computer Model Versus Empirical Methods

Authors: Kamil KAYABALI

Abstract: Soil liquefaction techniques based on in situ tests mostly utilize Seed's simplified procedure. In this method, the most critical input is the value of peak horizontal ground acceleration (a{_max}) on the ground surface. The preferred way for calculating the max a on the ground surface is either the use of statistical data from different soil conditions or the use of site response models. The max value of a is affected by both local soil conditions and nature of the base motion. In order to compare the shear stress values calmax culated using the Seed's simplitied procedure with the computer model-based shear stress values, ten sites with different soil thicknesses and shear wave velocities resulting in different soil periods were selected. The selected input motions also had different predominant periods. The computer program SHAKE91 was selected to compute the shear stresses at the pre-determined levels in the soil columns. It was found that Seed's approach overestimates the maximum shear stress induced by an earthquake in a soil profile by as much as 200% in some cases.

Keywords: Liquefaction, Seed's approach, shear stress, soil amplitication.