Oil spills pose a serious threat to marine ecosystems and coastal communities, and the risk in the Arctic is growing as traffic through the North-East Passage increases and fishing remains high. Vast distances, harsh conditions and long periods of darkness make detection and response difficult, while geopolitical tension can reduce transparency: some vessels may mask their position, which complicates emergency planning. Together, these factors create a clear need for faster, clearer situational awareness across agencies and operators.
Today’s monitoring often delivers static snapshots:
- Reports arrive as annotated images and files that represent a single moment, sometimes after delays of 20 minutes to several hours.
- Generic detection thresholds overlook local baselines.
- Forecast models may not assimilate real-time, in-situ data.
- Decision-makers are then left to act with partial information, increasing the chance of late or misdirected response.
This project addresses those gaps with a timeline service that integrates observations and predictions into one evolving picture. It will merge satellite data (SAR and optical) and it will use AI to detect and classify spills. A learned “normal state” for each area will help highlight anomalies quickly.
SINTEF will implement a service that provides the best available data for any location and time window, combining multiple sources and exposing a clear application programming interface for external systems. An Oil Drift Service will fuse several external drift simulators into a flexible ensemble; each run will describe its preconditions and uncertainties, giving responders better guidance for planning and action.
Header image: Shutterstock.