A Normalized Soil Parameter Procedure for Evaluating In-Situ Undrained Shear Strength

Abstract

Abstract The authors examined the results of 172 CKoU direct simple shear (DSS) tests. Their research has led to four important conclusions. Firstly, the test results confirm observations by other investigators that an acceptable correlation does not exist between the shear strength ratio, cu/?'v, and plasticity index, Ip. Secondly, consolidation pressure has a greater effect on cu/?'v than Ip and should be considered when evaluating DSS test results for use in a normalized soil parameter (NSP) procedure such as SHANSEP (stress history and normalized soil engineering properties). Thirdly, correlating cu/?'v results from DSS tests with only Ip could lead to either overestimating or underestimating in-situ undrained shear strength when employing an NSP procedure. Lastly, the authors demonstrate that correlations of DSS measured soil undrained shear strength, consolidation pressure and water content can provide a useful tool for evaluating in-situ undrained shear strength. Introduction Over the past thirty years, the DSS test has become widely used in geotechnical investigations, particularly in deepwater regions. The results of these tests are typically employed in some type of NSP procedure to evaluate in-situ undrained shear strength of clay deposits. The NSP procedure known as SHANSEP1 is a common method employed to perform this task. However, because of the expense and the considerable duration of testing associated with the SHANSEP procedure, laboratory testing is typically limited to a few tests within soil units defining the soil stratigraphy being evaluated. Subsequently, the results are often correlated with Ip To interpolate between SHANSEP test intervals to define the interpreted shear strength profile. A common correlation used to interpolate data between SHANSEP test intervals is that of the strength ratio, cu/?'v, and Ip. However, the authors' experience has been that a very poor correlation exists between cu/?'v and Ip when evaluated over a wide range of Ip common to offshore soils. This same opinion has been expressed by other investigators.2,3 To assess the reliability of the cu/?'v and Ip correlation, the authors examined the results of 172 DSS specimens tested in a normally consolidated state (OCR = 1). These tests were performed by five different geotechnical laboratories with extensive experience in performing SHANSEP-type testing. The soil specimens are from geotechnical site investigations conducted in six different offshore regions of the world. The plasticity characteristics of the 172 specimens examined in this study are presented in Fig. 1. This plasticity chart reveals that most of the specimens are highly plastic (CH) clays with liquid limits as high as 143 percent and Ip values as high as 101 percent. There are also several lower plasticity (CL) clays with Ip values ranging from 14 to 29 percent as well as some elastic silts (MH) and organic (OH) clays that fall below the A-line. The database does not include highly sensitive, cemented, or highly structured samples. Prevailing Strength Ratio - Ip Correlation In 1957, Skempton4 proposed the following correlation for normally consolidated clays based on field vane test results: (Mathematical equation available in full paper)