Oil spill risks associated with oil and gas exploration and production are changing as industry moves to deeper water. Computer-based modeling is typically used to quantify the fate and trajectory of sub sea releases (e.g., well blowouts). Stochastic modeling has become common as computational resources have grown, allowing for a large number of combinations of met ocean parameters to be evaluated. Simulating many potential realizations of a single discharge event allows for the calculation of spatially varying probabilities of oiling due to natural variability in the prevailing wind and current, but this needs to be treated separately from the likelihood of the sub sea release occurring to appropriately evaluate the oil spill risks of the offshore activity. Some industry guidance explicitly recognizes that the likelihood of a discharge event occurring is independent of the natural variability of the environmental setting and correctly treat these items as statistically independent; however, this distinction is sometimes lost in practice. In these instances, the resulting oil spill risk assessments may inadvertently be based on highly improbable combinations of release conditions and met ocean forcing.
A transparent and robust risk-based oil spill risk assessment needs to explicitly communicate the differences between the likelihood of a release event occurring and the probabilities of the potential trajectories that could result under different combinations of met ocean parameters. This poster uses a hypothetical deepwater blowout to illustrate how stochastic and deterministic modeling can be combined to characterize the probability distribution associated with the (variable) potential consequence of a discharge event. The proposed modeling strategy allows for the set of trajectories generated by stochastic modeling to be ranked based on various metrics of interest (e.g., volume of water column swept, area of surface slick, or volume of oil ashore) so the distribution of potential consequence can be evaluated. Following recently industry guidance (IPECA-OGP), it is suggested that the most probably deterministic trajectory be paired with statistical analysis, although the strategy allows for an identifiable amount of conservatism to be incorporated into the analyses by selecting one or more other trajectories for detailed impacts assessment.