Abstract

Damage to structures during earthquakes may be fully or partly caused by soil liquefaction, which has been the subject of extensive research for several decades. Liquefaction susceptibility of a sandy deposit is performed by comparing the resistance of a soil to liquefaction (i.e., capacity) to the load imparted by an earthquake (i.e., demand). In this regard, the stress-based method of liquefaction assessment is by far the most popular. It involves uncertainties mostly related to the computation of the maximum horizontal ground acceleration (amax) at bedrock. A site response analysis or a simplifi ed assumption is necessary to determine the amax on the ground level as well. Developing from the stress-based approach, the strain-based approach has also similar constraints. There exist laboratory techniques such as torsional shear to determine the capacity of a sandy soil in terms of liquefaction energy per unit volume. Likewise, the energy of a strong motion record can be set by employing simple physics principles. For this, a velocity time history and the unit mass of the soil are employed to compute the demand of any strong motion record. The scope of this investigation is to illustrate the usability of the energy-based method for the evaluation of soil liquefaction. The defi ciencies of the stress- and strain-based approaches are outlined and the advantages of the energybased approach are discussed.

Authors and Affiliations