Robb Eric S. Moss
Quantifying measurement uncertainty of thirty-meter shear-wave velocity
Bulletin of the Seismological Society of America (June 2008), 98(3):1399-1411

Abstract:
Thirty-meter shear-wave velocity (V _S30 ) is commonly used to estimate near-surface conditions for site classification, seismic site response, liquefaction analysis, and other geotechnical earthquake engineering application. Quantifying the uncertainty in V _S30 measurements is important for determining the accuracy and precision of this geophysical test. This study gathers existing available comparative and blind shear-wave velocity tests to evaluate the apparent or observable intra- and intermethod variability. Presented in this article are estimates of V _S30 uncertainty for the invasive methods of downhole, suspension logging, and seismic cone penetration testing, the noninvasive method of spectral analysis of surface waves, and the method of shear-wave velocity correlated geologic units. Discussed is the issue of soil disturbance with respect to invasive methods and how this may result in measurement bias. Results of this study indicate that uncertainty of shear-wave velocity measurements are on the order of 1%-3% coefficient of variation (standard deviation/mean) for downhole, suspension logging, and seismic cone penetration testing, 5%-6% coefficient of variation for spectral analysis of surface waves, and 20%-35% coefficient of variation for shear-wave velocity correlated geologic units. Presented here are procedures for propagating the uncertainty and/or bias in forward analyses.