Stochastic modeling for offshore risk assessment of pile refusal triggered by the presence of boulders

A 3D stochastic model is developed for the spatial distribution of seafloor and subsurface boulders. Boulder interception data from borings recovered from a European wind farm is presented and statistically analyzed. The proposed model is applied to simulate the boulder fields over the domain of an entire wind farm site and the simulated boulder fields were found to be consistent with field observations from boreholes (i.e., boulder interception by borehole), quantitatively validating the capability of the proposed method to reconstruct the boulder field based on data obtained from site investigation. A general relationship between the fraction of borehole interception and the fraction of boulder volume is defined and can be used for rapid estimation of the number of boulders at a site based on borehole data. This 3D boulder model can be easily incorporated into pile driving risk analysis to estimate the probability of pile refusal at a site or used to calculate the amount of necessary excavation work for shallowly embedded cable routes. Results calculated from a case study using the developed stochastic model and the boring data collected from the Beatrice wind farm show that the average pile refusal rate is 6.3% with significant dependence on the spatial location.