Selection of dispersion coefficients for use in lagrangian spill transport models to preserve underlying flow dynamics and transport barriers

Document Type

Conference Proceeding

Date of Original Version



Existing spill models typically use a Lagrangian based method (usually random walk) to model the transport of oil. The advective fields are most often obtained from a hydrodynamic model, but may sometimes come from high frequency radar (HFR) systems, or the analysis of satellite imagery. The dispersion coefficients required by the transport model are typically user selected based on the area of interest, but rarely come from hydrodynamic model predictions. If the selected horizontal dispersion coefficient is too high, transport of oil becomes dominated by dispersion and if too low the impact of turbulence on spill movement is under represented. In the former case, information about the current field and its structure, is masked by dispersive transport and the potential barriers to transport effectively eliminated. A methodology has been developed that allows estimates to be made of the upper bound for dispersion coefficients used in a spill model to ensure that barriers to spill transport are identified and accurately accounted for in the spill model. The relative dispersion of uniformly seeded Lagrangian trajectories is computed for increasing values of the dispersion coefficient until the mixing barrier is no longer effective. The dispersion coefficient, at which the mixing barrier disappears, provides a dynamical estimate of the upper bound of its value. The method has been tested using a simulation of the circulation for a few day period during the Deepwater Horizon spill period using results from the SABGOM hydrodynamic model hindcast of surface and subsurface currents.

Publication Title

39th AMOP Technical Seminar on Environmental Contamination and Response

This document is currently not available here.