Test Procedures for Measuring Soil Liquefaction Characteristics

The merits and limitations of laboratory simple shear and triaxial compression tests, conducted under cyclic loading conditions, are analyzed. In addition the theoretical relationships between the stress conditions causing liquefaction in simple shear and triaxial compression tests, based on several different failure criteria, are investigated. Finally data are presented for 35 sites where liquefaction is known to have occurred or not occurred during earthquakes, providing a means for evaluating the stress conditions causing liquefaction of sands in the field. It is shown that both analytical and experimental evidence, as well as field data, suggest that most laboratory triaxial compression test data, expressed in terms of σ_dc/(2σ_a) should be reduced by a factor of 0.55 to 0.7, depending on the density of the soil and the duration of the earthquake, in order to determine values of τDhv/σ'_o causing liquefaction during earthquakes under field conditions. Considerable judgment is involved in evaluating such relationships, and data from field studies may provide the basis for evaluating the liquefaction characteristics of sands.