Implications from a geotechnical investigation of liquefaction phenomena associated with seismic events in the Charleston, SC area

First-hand accounts of sand boils and other liquefaction-related phenomena associated with the Charleston, SC earthquake of 1886 provide clear evidence that liquefaction was common in this event. Recent geologic investigations in the Charleston area have found evidence for the repeated liquefaction of sandy soils in the Charleston area due to recurring large seismic events. Although this information has led to an improved understanding of seismicity in the Charleston region, little hard data exists in terms of ground motion characteristics or levels of seismic loading. A two-year field investigation was undertaken by Virginia Tech to study the liquefaction findings associated with the 1886 event from the perspective of geotechnical engineering. This involved defining the engineering parameters of the Charleston soils on the basis of in-situ and laboratory tests, and estimating the levels of seismic loading required to produce the observed liquefaction phenomena.

Of the sites where field tests were performed, the surficial soils were largely formed from ancient beach ridge deposits. The findings showed that soil conditions within these deposits are appropriate for liquefaction. Also, there is clear evidence that soils as old as 230,000 years have liquefied multiple times in the past 10,000 years. Many of these soils remain susceptible to liquefaction at relatively low levels of seismic shaking, although there is some evidence for progressive densification.

With respect to the seismic loadings, evidence is presented which suggests that both the magnitude and peak acceleration of the 1886 earthquake were less than what has been proposed by the seismological community (M = 7.7 and 0.5 - 0.6g peak acceleration). The findings of this study indicate that for an M = 7.5 event, peak accelerations in the 0.3 to 0.4g range would serve to explain the observed 1886 liquefaction phenomena. If it is assumed that the magnitude of the 1886 earthquake was less than 7.5, then the estimated peak accelerations increase.