Trajectory Forecast as a Rapid Response to the Deepwater Horizon Oil Spill

In response to the Gulf of Mexico Deepwater Horizon oil spill, a Lagrangiantrajectory modeling system was implemented immediately upon spill onset by marshaling numerical model and satellite remote sensing resources available from existing coastal ocean-observing activities. Surface oil locations inferred from satellite imagery were used to re-initialize the positions of virtual particles in this ensemble of trajectory models, and the particles were tracked using forecast surface currents, with new particles added to simulate the continual release of oil from the well. A challenge to this modeling effort was that much information remained unknown throughout the spill event, with additional uncertainty due to intensive mitigation activities. By frequently re-initializing the trajectory models with satelliteinferred locations, the effects of in situ mitigation and forecast error growth were implicitly accounted for and minimized. The simulated surface oil trajectories were compared to the satellite observations in subsequent forecast cycles for veracity testing. Although similar results were obtained, in general, differences were seen in the simulated trajectories by different models. However, no one model performed consistently better or worse than the others throughout the event with one exception. The lessons learned from the event may be useful in preparing rapid trajectory forecast systems in the future.