| Though soil liquefaction is commonly associated with loose sands, silt has been known to liquefy during large earthquakes. Unlike sands, however, the cyclic resistance of silt is difficult to quantify because current semi-empirical correlations are based largely on sandy soils having a fines content of less than about 35%. To better understand the cyclic resistance of silt, a cyclic triaxial testing program was conducted on non-plastic silts recovered from various sites in the Providence, RI area. A new modified moist tamping method was introduced for preparing reconstituted samples of silt for liquefaction testing. By comparing the results to undisturbed block samples, the new moist tamping method was shown to capture the in situ behavior up to the initiation of liquefaction. Correlations were developed for two Providence silts that relate cyclic resistance to shear wave velocity (VS ). One of the correlations was used to estimate the in situ cyclic resistance of a silt deposit at a study site using in situ VS measurements made with a seismic cone. The cyclic resistance estimated from the new correlation compared well to the resistance predicted using current Standard Penetration Test (SPT) and Cone Penetration Test (CPT)-based methods. An analysis of liquefaction potential was also performed at the study site. The seismic loading on the subsurface soils was estimated from ground response analyses using actual recorded earthquake time histories. The analysis indicates the potential for liquefaction within a depth interval of 5.5 to 7.5 m below the ground surface in a 2,500-year design level earthquake. |
