Abstract: Uncertainty in geotechnical engineering greatly results in engineering risk. In this study, the first large-scale uncertainty experiment on the in-situ shear-wave velocity (Vs) was carried out on 20 typical sites in 7 major regions of China. A total of 11 kinds of common engineering instruments were completed by using single hole method in 47 units, and 600 groups of Vs were obtained finally. According to this in-situ database, the corresponding formula between Coefficient of variation (COV) of time-average shear-wave velocity (Vs, z) and the calculated depth (Z) was fitted. We revealed the impact of in-situ Vs uncertainty on the judgement of site stiffness and pointed out the possible potential misjudging area of site classification in China. At the current level of in-situ Vs test in China, COV of Vs, z is negatively correlated with depth, which indicates that it reaches the max value of 15% near the site surface and then decreases significantly with the increasing of the depth. In the depth range from 0 m to 10 m, the COV of Vs, z decreases rapidly; in the depth range of 10-20 m, the COV of Vs, z decreases slowly; after the depth of 20 m, the COV of Vs, z is less than 5% and basically unchanged; the COV of equivalent shear wave velocity (Vse) used for site classification of Class III and IV sites in China is basically the same as that of the field classification index Vs, 30 used in Europe and America. Based on China building code GB50011-2010, Class I and Class II sites with the soil thickness less than 5 m or Vse close to 500 m/s should consider the potential impact of Vs uncertainty, while other types of Class II and Class III and Class IV sites could ignore the uncertainty.

Key words: shear-wave velocity, uncertainty, site stiffness, Time-average Shear-wave Velocity