Probabilistic analysis of offshore geotechnical site investigation in a homogeneous stiff clay deposit

Abstract

Unlike most onshore geotechnical projects, conducting offshore site investigations is often difficult and technically challenging due to the harsh ocean environment. Hence, an important cost consideration is minimizing the number of soundings for an offshore project while ensuring reliable and safe designs. This study examines the possibility of reducing the number of soundings at site by using practical numerical tools to predict cone tip resistance (qt ) at unsampled locations. Prediction errors were quantified within a probabilistic framework. Two prediction approaches, i.e. 2D linear regression and Kriging, were adopted. Site investigation data from an existing offshore wind farm with homogeneous stiff clay deposit were collected and analyzed. Result shows that in a homogeneous soil deposit, it is possible to use as little as 20% of the available data to predict the remaining 80% qt profiles with small error. Monopile deformation under typical design load was also calculated using shear strength derived from the qt profile at each wind turbine location. The small spatial variation of monopile deformation suggests that a reduction in the number of geotechnical soundings for sites with uniform soil conditions would not significantly impact the foundation design.