中文摘要：

剪切波速也正逐步成为土层液化判别的基本指标之一,但采用现场波速资料得到的砂砾土液化判别方法尚较少见。针对2008年汶川8.0级地震显著的砂砾土液化现象,获取45个场地剪切波速结构,以此提出基于剪切波速的砂砾土液化判别方法;构建相应模型和计算公式,并分析现有2种典型砂土液化剪切波速判别方法对砂砾土的适用性。提出的砂砾土液化剪切波速判别方法由初判和复判组成,初判包括地质年代、埋藏条件和含砾量3个条件;复判模型则由地震烈度、剪切波速基准值、地下水位、砂砾土埋深和和含砾量等5个参数构成,并分别采用归一化方法和优化方法推导出剪切波速基准值以及地下水位和砂砾土埋深的影响系数。砂砾土与砂土属不同土类,相同波速值下二者密实程度不同,现有砂土液化剪切波速判别方法对砂砾土不适用,给出的判别结果明显偏于危险。获取的砂砾土液化资料扩充现有液化数据库内容,提出的砂砾土液化剪切波速判别方法简单明了,回判成功率高,可为工程应用及规范修订提供参考。

英文摘要：

Although the in-situ shear wave velocity tests have been used commonly as engineering testing techniques and shear wave velocity Vs has gradually become a basic index for soil liquefaction evaluation,the in-situ Vs-based liquefaction assessment methods for gravel soils are not available so far. The significant phenomena of gravelly soil liquefaction in the 2008 Wenchuan 8.0 Earthquake are investigated and the in-situ Vs structures for 45 sites in the event are obtained. In terms of the investigation data,the Vs-based approach for liquefaction evaluation of gravel soils is presented and the existing Vs-based methods for evaluating sandy soil liquefaction are inspected. The liquefaction discrimination of gravel soils can be divided into the two steps：the preliminary evaluation and further evaluation. In the preliminary step the geological ages,the buried condition of gravel soil layer and gravel contents of gravel soils are considered and five parameters including seismic intensity,reference values of Vs with depth of gravel soils,groundwater table and gravel content of gravel soils are concerned in the further step. The reference values of Vs as well as the influential coefficients of gravel soil depths and groundwater levels are deduced by normalization method and optimization method respectively. As the type of gravel soils differs from the sandy soils,the densities of two type soils even for the same Vs value are different. If the existing liquefaction assessment procedures for sandy soils are employed,the liquefaction resistance ofgravel soils will be significantly overestimated,which means that the methods of liquefaction evaluation for sandy soils are not suitable for gravel soils. The proposed method is noticed with clear expression and high success ratio of regression discrimination;and it can easily be used in engineering practice and code revision.