Rising Arctic maritime activities and hydrocarbon development increase the risk of an oil spill in and under Arctic sea-ice. Oil spilled under growing sea ice would be encapsulated within the ice cover. During spring and early summer, such trapped oil would migrate upwards, pervading the ice volume and ultimately pooling at the surface. Current gaps in our understanding of these processes have major implications for spill clean-up efforts and habitat damage assessments. Guided by results from three sets of ice-tank experiments, we have
developed a semi-empirical multi-stage oil migration and surfacing model to help predict oil in ice behavior relevant to spill response. According to previous studies, upon under-ice release, oil saturates the ice skeletal layer, remaining largely immobile during the growth season. As intrinsic ice permeability increases above 10-11 m2 with the onset of surface melt, oil migrates rapidly through the full depth of the ice cover, primarily through the secondary pore space. Then, the ever-increasing connectivity between the pores allows the oil to invade the primary pore space. Finally, as the ice deteriorates, oil occupies most of the pore space. Our stratigraphic analysis revealed that granular ice impedes surfacing of oil in cold ice due to the more tortuous pore space. It also showed that the potential for oil movement during the growth season is constrained by the availability of migration pathways from the oil/ice interface to the surface. In contrast to previous findings, our results indicate that if such oil migration pathways are present, significant oil mobilization can occur in cold ice during the growth season. Thus, we tracked upward oil migration through large brine channels in cold ice (Tice <–5 °C) over vertical
distances of up to 30 cm, leading to surfacing of oil, within 24 hours after release. During ice melt and deterioration, oil movement is tied to the magnitude of the bulk brine volume fraction and the magnitude of the oil lens reservoir. Development of a predictive oil migration model based on these findings will aid spill response planning, oil detection and damage assessments.