Chinese code methods for liquefaction potential assessment based on standard penetration test: An extension

Highlights

• The depth consistency assumption embedded in the Chinese code methods is assessed.

• A new model is developed to overcome the limitations of the Chinese code methods.

• The suggested method can be used for liquefaction assessment internationally.

• The performance of the suggested method is compared with four representative methods.

• The suggested method provides a useful alternative for liquefaction assessment.

Abstract

In China, the required blow count is used to represent the effect of seismic loading, which can provide engineers an intuitive feeling about how much effort is needed to improve the soil if its liquefaction resistance is not adequate. A fundamental assumption in the Chinese seismic design code methods is depth consistency, i.e., the required blow count is a non-decreasing function of the depth. Nevertheless, the methods described in the Chinese seismic design codes are subjected to several limitations, and they are mainly used within China. In this paper, the validity of the depth consistency assumption is discussed, and a new method is developed based on the concept of depth consistency that can overcome the shortcomings of the methods described in the Chinese seismic design codes and hence can be used internationally. For the Cetin et al. [1] database and Chi-Chi earthquake database, it is shown that the method suggested in this paper is more accurate than the method described in the current Chinese seismic design code. Also, it can perform as well as and, in some cases, even outperform representative SPT-based empirical methods built within the cyclic stress ratio (CSR) framework. An illustrative example is presented to illustrate the application of the suggested method.

Introduction

Earthquake-induced liquefaction in saturated sand deposits may contribute to ground failures including sand boiling, foundation settlements, and lateral spreading, etc., which can cause serious damage to infrastructure and lifelines [2,3]. In practice, empirical methods based on in-situ tests such as the standard penetration test (SPT) [[4], [5], [6], [7], [8]], the cone penetration test (CPT) [[9], [10], [11], [12], [13]], and the shear-wave velocity (V_s) [[14], [15], [16]] have been widely used to assess the liquefaction potential of soils.

In China, the SPT-based method is most widely used. The SPT-based method for liquefaction assessment was initially established in the Chinese seismic design code in 1974 [17] based on the review of case histories collected from several earthquakes that occurred in China with reference to pioneering studies by Seed and Idriss [18], and has been revised several times in different versions of the seismic design codes considering the case histories from the 1975 Haicheng earthquake and 1976 Tangshan earthquake [[19], [20], [21]]. In this study, these methods are collectively called the Chinese code methods as they are specified in different versions of the Chinese seismic design codes, and are widely used in China. In the Chinese code methods, the seismic loading is represented by the required uncorrected blow count (N_req), which is different from that in the empirical methods built based on the cyclic stress ratio (CSR) as explored by Seed and his colleagues [6,18]. For example, in the method specified in the current Chinese seismic design code [20], which is referred to hereafter as MOHURD2016, N_req is calculated as follows:

$$N_{\text{req}}=N_0\beta \left[\ln \left(0.6d_s+1.5\right)-0.1d_w\right]\sqrt{\frac{3}{{\rho }_c}}\left(d_s\le 20\text{ m}\right)$$

$$\beta =0.25M-0.89$$

where d_s is the depth of soil being assessed in meters; d_w is the depth of groundwater table below ground surface in meters; ρ_c is the clay content (particle size smaller than 0.005 mm) in percentage, which should be set to 3 if it is less than 3 or the soil type is sandy soil; β is the adjustment factor considering the influence of earthquake magnitude; M is the earthquake magnitude without specifying the type of magnitude scale, while moment magnitude (M_w) is preferred; N₀ is the reference SPT blow count for the case of M = 7.5, d_s = 3 m and d_w = 2 m and probability of liquefaction (P_L) = 0.32, and its value is determined according to the horizontal peak ground acceleration a_max (g) as shown in Table 1, where g is the acceleration of gravity. In the Chinese code methods, the uncorrected SPT blow count (N) is adopted to represent the liquefaction resistance of soils. The soil is likely to liquefy if N_req > N, and vice versa. The Chinese code methods provide engineers an intuitive feeling about how much effort is needed to improve the soil in terms of the SPT blow count if its liquefaction resistance is not adequate.

A fundamental assumption in the Chinese code methods is that the required blow count is a non-decreasing function of the depth. Such an assumption is called depth consistency in Yang et al. [22]. The Chinese code methods provide an alternative point of view on how empirical methods can be built based on case histories. While these methods have been used in China for many years, there could be several limitations when they are used internationally: (1) the case histories used for establishing MOHURD2016 were collected from earthquakes that occurred in China; (2) the uncorrected blow count is used for liquefaction potential assessment, which does not account for the corrections of hammer energy efficiency and effective overburden stress; (3) the calibration method of model parameters is not mathematically rigorous; and (4) it is difficult to plot all the case histories which are associated with different values of M_w, a_max and d_w in the 2-D space of N - d_s to inspect the accuracy of MOHURD2016, because the N_req - d_s triggering curve is dependent on M_w, a_max and d_w. For comparison, it is convenient and common to plot all case histories in a 2-D graph to inspect the model performance visually within the cyclic stress ratio (CSR) framework.

Probably because of the above limitations, the Chinese code methods are mainly used within China. The purpose of this study is to develop a liquefaction potential assessment model following the depth consistency assumption embedded in the Chinese code methods, which can effectively overcome the above limitations of the Chinese code methods and can hence be used internationally. This paper is organized as follows. First, the depth consistency assumption is assessed through models built within the CSR framework. Then, the suggested method for liquefaction potential assessment is described. Thereafter, the prediction accuracy of the suggested method is compared with four representative methods in the literature. Finally, a procedure for presenting the suggested method in a 2-D plot with all case histories is described, and an illustrative example is used to demonstrate the suggested method for liquefaction potential assessment.