Expert elicitation of directional metocean parameters

Abstract

Probability distributions that describe metocean conditions are essential for design and operational decision making in offshore engineering. When data are insufficient to estimate these distributions an alternative is expert elicitation -a collection of techniques that translate personal qualitative knowledge into subjective probability distributions. We discuss elicitation of surface currents on the Exmouth Plateau, North-Western Australia, a region of intense oil and gas drilling and exploration. Metocean and offshore engineering experts agree that surface currents on the plateau exhibit large spatio-temporal variation, and that recorded observations do not fully capture this variability. Combining such experts' knowledge, we elicit the joint distribution of magnitude and direction by first focusing on the marginal distribution of direction, followed by the conditional distribution of magnitude given direction. Although we focus on surface currents, the direction/magnitude components are common to many metocean processes. The directional component complicates the problem by introducing circular probability distributions. The subjectivity of elicitation demands caution and transparency, and this is addressed by embedding our method into the established elicitation protocol, the Sheffield Elicitation Framework. The result is a general framework for eliciting metocean conditions when data are insufficient to estimate probabilistic summaries.

Emulation of vessel motion simulators for computationally efficient uncertainty quantification

“…This exploits the inherent correlated structure present in the inputs, allowing us to obtain random draws from the joint distribution of wind, wind-wave, and swell. The elicitation results in Astfalck et al (2018) present seasonal distributions for surface current magnitude and direction, at the location of the FPSO. We augment the training samples from the hindcast dataset with seasonally matched samples from the elicited distributions of surface current.…”

“…Table 1 lists the considered inputs; this results in a ten dimensional input space with both circular and linear variables. A numerical hindcast dataset is used to inform wind, wind-wave, and swell parameters, and the results of an expert elicitation workshop are used to inform surface current parameters (Astfalck et al, 2018).…”

“…Hindcast data of surface currents were however deemed to not be reliable, owing to their particularly high spatial and temporal variability. In lieu of field measurement of the currents, results of a recent expert elicitation of the surface currents at the site were utilised (Astfalck et al, 2018). Severe tropical cyclones are expected during the Wet season, during which the vessel is expected to disconnect.…”

“…Unfortunately, surface current hindcasts were not available over this time span. Instead, we augment the hindcast dataset with samples from the distributional results of a recent expert elicitation workshop on surface currents on the North-West Shelf's Exmouth Plateau as described in Astfalck et al (2018). It is vital to confirm the emulator is a reasonable surrogate of the simulator and the statistical diagnostics presented in Bastos and O'Hagan (2009) are used to assess the emulator performance.…”

Uncertainty Quantification of Density and Stratification Estimates with Implications for Predicting Ocean Dynamics

“…To complete the Bayesian specification of the model, we require prior distributions on all remaining model parameters. The following priors for the regression coefficient means are the result of an expert elicitation process (Astfalck et al 2018) about the possible behavior of vertical density profiles. Given the physical interpretation of these parameters, as discussed in section 3 and shown in Fig.…”