Oil Spill Amount Estimation Method Based on an Adjoint Assimilation Model: A Case Study of Penglai 19–3 Oil Spill

Abstract Accurate estimation of oil amounts is crucial for effectively assessing the risks associated with marine oil spills, necessitating a reliance on effective monitoring, numerical models, and statistical analysis. In this study, an oil transport and assimilation model is developed for the Bohai Sea to assess the total amount of oil spill from the Penglai 19–3 oil spill incident based on the oil degradation experiments. By assimilating observation data in twin experiments, it is demonstrated that Cressman interpolation can improve the estimated initial oil concentration as well as significantly reduce the mean absolute error (MAE) between the observed and simulated results. In practical experiments, considering the oil degradation process, the total amount of spilled oil was calculated to be 8203.32 tonnes, with the maximum expected oil loss being 3511.27 tonnes. The results are helpful for informing oil spill damage assessment and guiding emergency treatment. Significance Statement To our knowledge, this is the first time to describe oil transport characteristics of the semienclosed continental shelf sea area, the Bohai Sea, using an adjoint assimilation model. We parameterized the experimental oil degradation term in the conventional convection–diffusion equations and used an adjoint model to estimate the total amount of spilled oil. The results show that a total of 8203.32 tonnes of oil were released into the sea, contaminating approximately 4782 km 2 of coastal water, which is about 7% of the entire Bohai Sea area. Notably, the model can estimate the amount of oil loss during the degradation process.