A Dynamic and Quantitative Risk Assessment Method for Marine Oil Spills Accruing in Extreme Winds Based on Dynamic Bayesian Network

Zengkai Liu,Zhonghao Han,Qi Chen,Xuewei Shi,Qiang Ma,Baoping Cai,Yonghong Liu
DOI: https://doi.org/10.2139/ssrn.4158226
2022-01-01
SSRN Electronic Journal
Abstract:Oil spills are serious threats to the marine ecosystem. Especially when an oil spill is faced with extreme weather, the consequences might be more severe. Until now, no such researches focus on oil spill risk occurring in annual extreme winds. This paper proposes a novel dynamic assessment method to quantify the risk of oil spill in extreme winds based on dynamic Bayesian network (DBN). In this paper, the physical models of advection, spreading, evaporation, and dispersion are transformed into DBNs, and the Bayesian network of vulnerability is established according to coastline types and socio-economic resources. By integrating all the sub-models, the overall DBN to quantify the dynamic risk of oil spills occurring in extreme winds is obtained. The proposed method is demonstrated by Laizhou Bay. The developed model is validated by a three-axiom-based approach. Temporal and spatial dynamics of risk caused by oil spills in potential locations and driven by annual extreme winds could be calculated. Based on the developed DBN, risk of Laizhou Bay coast caused by oil spills in annual extreme wind speeds of different mean recurrence intervals are researched. In addition, the effects of occurrence time of annual extreme winds on risk of the coast are researched.
What problem does this paper attempt to address?