Estimating relative density from shallow depth CPTs in normally consolidated and overconsolidated siliceous sand

Abstract

Current interpretation of relative density ( Dr) based on cone penetration test (CPT) end resistance ( qc) data in sand relies on empirical expressions established from calibration chamber studies for which a deep failure penetration mechanism is attained. These expressions are mainly based on stress normalised qc data. Previous studies have highlighted the importance of performing the normalisation procedure with respect to the mean effective stress ( p′) in overconsolidated (OC) sand deposits instead of the vertical effective stress ( σ′v) that has been used in normally consolidated deposits. Due to a large variation in the coefficient of earth pressure at rest ( K0), on which p′ depends, in the uppermost 3–5 m of OC sand, a recent study has demonstrated a rigorous unified approach for estimating a varying K0 with depth. However, the surficial shallow failure penetration effects are not accounted for, and consequently, interpretation of the upper 0.5–1.5 m of sand is still erroneous. This depth range is very important for low-stress geotechnical applications such as offshore flowlines and mudmats. With a reinterpretation of previously published data, a new global model is presented that potentially enables estimation of Dr from shallow CPTs in siliceous sand by taking into account both the shallow failure penetration effects (important in the uppermost 0.5–1.5 m) as well as the varying K0 with depth for OC sand (important in the uppermost 3–5 m). However, the new model needs future calibration with controlled laboratory testing.