An evaluation of models that estimate droplet size from subsurface oil releases

Abstract

Droplet size substantially affects the fate of oil released from deep subsea leaks. A baseline dataset of volume-median droplet diameters (d₅₀), culled from ~250 laboratory observations, is used to validate seven droplet-size models. Four models compare reasonably well, having 95% confidence limits in d₅₀ of ~±50%. Simulations with a near-field fate model (TAMOC) reveals that the four best-performing models, with d₅₀ of 1.3-2.2 mm, agree similarly with observed fractionation of petroleum compounds in the water column during June 4-July 15, 2010. Model results suggest that, had a higher dose of dispersant been applied at the wellhead during Deepwater Horizon oil spill (DWH), the d₅₀ would have dropped by an order of magnitude, reducing surfacing C₁-C₉ volatiles by 3.5×. Model uncertainty is found to be substantial for DWH-like blowouts treated with chemical dispersants, suggesting the need for further droplet-size model improvement.

Keywords: Blowouts; Oil droplet sizes; Oil pipeline leaks; Oil spill models; Oil spills.