Reliability-Based Assessment of Liquefaction Potential Using SPT Approach

Earthquake-induced liquefaction is one of the most complex and interesting phenomenon in Geotechnical Earthquake Engineering. Liquefaction is a very significant phenomenon in alluvial soil deposits consisting of silty sand or sandy silt type of soils. Many different methods are available at present to assess the liquefaction potential of soil using in situ field test data. Generally, field tests like SPT are carried out and liquefaction assessment is done for each borehole. Different boreholes provide different depths up to which soil has potential for liquefaction for particular peak ground acceleration (PGA) and magnitude of Earthquake. Maximum liquefaction depth obtained from different boreholes is used to convey the liquefaction depth of the area in consideration. The greatest drawback of this approach is that results of only one borehole are used to conservatively predict the liquefaction potential of the entire area and results of remaining boreholes are neglected as they have predicted lesser depth of liquefaction. In case some more boreholes are drilled and assessment is done then there is quite a good probability that liquefaction depth of that area will change depending upon the SPT results from new boreholes. In this paper liquefaction potential assessment of alluvial soil site is carried out using SPT-based approach as proposed by NCEER [5]. SPT data of 25 boreholes are analyzed and converted into the equivalent single borehole using probabilistic approach to assess the liquefaction potential of alluvial soil site. Probabilistic approach is used to assess 95 percentile values of all variables required to assess Liquefaction potential like SPT blow counts, percentage fines, and soil density. As there is uncertainty present in the evaluation of these parameters at the site and in the laboratory, it is appropriate to evaluate these parameters based on a probabilistic approach using the best fit probabilistic distribution curve. Parameters required for assessing CSR (Cyclic Stress Ratio) like overburden pressure and effective overburden pressure are also analyzed using probabilistic distribution curves considering data from 25 boreholes and converting them into an equivalent single borehole. Factor of Safety (FOS) obtained using 50, 95, and 98 percentile values of different parameters is compared for converted equivalent single borehole.