Assessment and development of empirical correlations between flat plate dilatometer and cone penetration tests for Auckland soils

Abstract

ABSTRACT Based on a database of over 2900 cone penetration test (CPT) and flat plate dilatometer test (DMT) data pairs obtained from Auckland, New Zealand, this paper assesses the applicability of several empirical CPT-DMT correlations and presents newly developed regional correlations for use in geotechnical engineering applications. Empirical correlations using the DMT-based material index are assessed. As both the DMT-based material index (ID ) and the soil behaviour type index (Ic ) from CPT data are often used to estimate soil behaviour characteristics when no samples are retrieved, these were used to develop subset databases. In general, the correlation used to estimate the material index (ID ) tended to slightly overestimate the measured values for Auckland soils, while better performance could be observed after removing data points where Ic and ID soil classifications were incompatible. In contrast, the correlations used to estimate horizontal stress index (KD ), and dilatometer modulus (ED ) underestimated the measured values for the database used herein, with the largest underestimates for low cone tip resistance values. This paper presents new correlations for estimating KD and ED based on CPT parameters for Auckland soils, providing improved estimates, with stable bias values across the range of input parameters considered.