A Bayesian Network risk model for assessing oil spill recovery effectiveness in the ice-covered Northern Baltic Sea

• Published in: Marine Pollution Bulletin
• Main Authors: Lu, Liangliang, Goerlandt, Floris, Valdez Banda, Osiris, Kujala, Pentti, Höglund, Anders, Arneborg, Lars
• Other Authors: Department of Mechanical Engineering, Marine Technology, Swedish Meteorological and Hydrological Institute, Aalto-yliopisto, Aalto University
• Format: Article in Journal/Newspaper
• Language: English
• Published: PERGAMON-ELSEVIER SCIENCE LTD 2019
• Subjects: Risk analysis; Oil spill; Recovery effectiveness; Sea ice; Maritime safety; Marine pollution; ice covered waters
• Online Access: https://aaltodoc.aalto.fi/handle/123456789/36809; https://doi.org/10.1016/j.marpolbul.2018.12.018

The Northern Baltic Sea, as one of the few areas with busy ship traffic in ice-covered waters, is a typical sea area exposed to risk of ship accidents and oil spills in ice conditions. Therefore, oil spill capability for response and recovery in this area is required to reduce potential oil spill effects. Currently, there are no integrated, scenario-based models for oil spill response and recovery in ice conditions. This paper presents a Bayesian Network (BN) model for assessing oil spill recovery effectiveness, focusing on mechanical recovery. It aims to generate holistic understanding and insights about the oil spill-to-recovery phase, and to estimate oil recovery effectiveness in representative winter conditions. A number of test scenarios are shown and compared to get insight into the impact resulting from different oil types, spill sizes and winter conditions. The strength of evidence of the model is assessed in line with the adopted risk perspective. Peer reviewed