Oil Spill Modeling in Sea Ice Covered Ocean

Abstract

The ongoing reduction in extent and thickness of sea ice in the Arctic allows the expansion of shipping activity and oil exploration in the high north, and with that a potential increased risk of oil spill in ice covered areas. This thesis asses the response of two oil-in-ice surface drift models implemented in an open-source Lagrangian framework and forced by four dif- ferent ice-ocean products (RIOPS, TOPAZ4 real-time forecast system, TOPAZ4 reanalysis and SVIM). Both approaches were evaluated over three sets of simulations: (I) a field experiment conducted in the Barents Sea marginal ice zone in 2009; (II) observed trajectories of buoys in the ice pack and in the Barents Sea marginal ice zone; and (III) stochastic simulations (960 runs, from 1998 to 2017) to reproduce a hypothetical oil spill in the Kara Sea. Results from experiments (I) and (II) indicate that the two drift models provide similar response both in the ice pack and the marginal ice zone under the same forcing. It was also found that finer horizontal resolution ice-ocean products (RIOPS and SVIM) did not reproduce better the ob- served drifts. The experiment (III) revealed that the sea ice concentration (%) field dictates the spread, the predominant direction of trajectories and the distance (km) traveled by the cloud of particles (SVIM: -1.41 km/% and TOPAZ4 reanalysis: -1.24 km/%).