Uncertainty quantification and reliability assessment in operational oil spill forecast modeling system

Factor diagnosis and governance strategies of ship oil spill accidents based on formal concept analysis

Ship oil spill accidents have a prolonged duration, complex consequences, challenging cleaning and repairing efforts, and pose a significant threat to the environment, economy, and society. Eliminating irrelevant information and identifying key factors using traditional methods is challenging due to the complexity of the causes of ship oil spill accidents. To address this, this article sorts out the accident databases of the International Tanker Owners Pollution Federation (ITOPF) and eight national maritime administration agencies, and innovatively constructs a formal concept analysis (FCA) model based on reports of 100-plus ship oil spill accidents. The model results prove that improper operation, less complete ship equipment, large tonnage, and poor navigation conditions are the key factors. The different causal rules of oil spills in collision/contact, grounding, fire/explosion, and foundering are further compared and analyzed. Finally, corresponding improvement measures are put forward for the key factors of oil spills and different causal rules.

Numerical Study on the Influence of Model Uncertainties on the Transport of Underwater Spilled Oil

Oil pollution influences marine biology, ecology, and regional sustainable development capacity, but model uncertainties limit the ability of the numerical model to accurately predict the transport and fate of the underwater oil spill. Based on a three-dimensional underwater oil spill model validated by satellite images of the oil slick at the sea surface, the Penglai 19-3 oil spill accident in the Bohai Sea was simulated; in addition, several sensitivity experiments were set up to investigate the influence of model uncertainties in the background wind, current, start time of the oil spill, and spill site on the transport of underwater spilled oil in the Penglai 19-3 oil spill accident. The experimental results indicate that the uncertainty in the background wind has a certain impact on the simulated centroid position at the sea surface, and little effect on the simulated underwater results, while the uncertainty in the background current has a significant influence on the transport of the underwater spilled oil both at the sea surface and underwater. An uncertainty of 24 h in the start time of the oil spill can cause more than 1 time larger than the benchmark case displacement of the oil spill centroid point and sweeping area at the sea surface, as the periodic tidal current is the main constituent of the ocean current in the Bohai Sea. The uncertainty in the spill site has a large influence on the final position of the oil spill centroid point, but the oil spill trajectories do not intersect with each other within 48 h, which makes it possible to identify the oil spill platform from the actual observations. The influence of uncertainties in the important model inputs and key model parameters on the transport of underwater spilled oil in the Penglai 19-3 oil spill accident is evaluated for the first time, which is of substantial significance for improving the prediction accuracy of the transport and fate of underwater oil spills.

The oil spill transport across the shelf-estuary interface

Oil spills caused by ship collisions and offshore oil wells are an ongoing risk for estuaries in the northern Gulf of Mexico. The fate and transport of the oil spill across the interface between a bar-built estuary and the adjacent coast are influenced by multi-scale forcing mechanisms and their corresponding interactions. Of primary interest are the alongshore currents on the shelf encountering strong tidal flows at the estuary entrance. A new cross-scale model was developed for Galveston Bay to reproduce the multi-scale flows. The model was employed in regionally-distributed numerical Lagrangian experiments to investigate the oil spill transport across the shelf-estuary interface. The influence of the multi-scale flows on the oil spill transport was characterized in terms of Lagrangian connectivity and Lagrangian flushing. The new Galveston Bay model was also used to evaluate the Texas City "Y" spill and resulted in a reasonable agreement with the NOAA observations. This research enhances our understanding of the oil transport across the threshold between two contiguous water systems and highlights the importance of resolving the multi-scale flows for the purpose of oil spill predictions.

Innovative Approaches for Geometric Uncertainty Quantification in an Operational Oil Spill Modeling System

Reliable and rapid real-time prediction of likely oil transport paths is critical for decision-making from emergency response managers and timely clean-up after a spill. As high-resolution hydrodynamic models are slow, operational oil spill systems generally rely on relatively coarse-grid models to provide quick estimates of the near-future surface-water velocities and oil transport paths. However, the coarse grid resolution introduces model structural errors, which have been called “geometric uncertainty”. Presently, emergency response managers do not have readily-available methods for estimating how geometric uncertainty might affect predictions. This research develops new methods to quantify geometric uncertainty using fine- and coarse-grid models within a lagoonal estuary along the coast of the northern Gulf of Mexico. Using measures of geometric uncertainty, we propose and test a new data-driven uncertainty model along with a multi-model integration approach to quantify this uncertainty in an operational context. The data-driven uncertainty model is developed from a machine learning algorithm that provides a priori assessment of the prediction’s confidence degree. The multi-model integration generates ensemble predictions through comparison with limited fine-grid predictions. The two approaches provide explicit information on the expected scale of modeling errors induced by geometric uncertainty in a manner suitable for operational modeling.

Evaluation of the diagnostic ratios of adamantanes for identifying seriously weathered spilled oils from simulated experiment and actual oil spills

The composition and physical properties of spilled oil have great changes during the seriously weathering process. It brings great difficulties to the source identification of oil spill. So the stable and trustworthy diagnostic ratios (DRs) for accurate identification of severely weathered spilled oils are very important. The explosion of Sinopec pipeline happened on November 22, 2013 at Qingdao, China. Local beaches at Jiaozhou Bay were polluted by spilled oils. We have collected original spilled oil samples from an area free from human interference near the oil leakage point after the accident. Synchronized with actual beach weathering, laboratory experiments were conducted to simulate oil weathering for 360 days by using the collected original spilled oil samples. Based on t test and the repeatability limit method, 50 diagnostic ratios (DRs) of adamantanes were screened. Four DRs, namely 1,3-dimethyladamantane/total dimethyladamantane, 1-methyladamantane/(1-methyladamantane + 1,3-dimethyladamantane), dialkyl diamantane/total diamantane, and diamantane/(diamantane + dialkyl diamantane), have maintained remarkable stability during the simulated weathering experiments and field weathering process. These stable ratios can retain the characteristics of oil source during weathering. They are very beneficial to improve the accuracy of identifying the source of severely weathered oil and can be used as an effective supplement to existing index system for source identification.

Performance evaluation of oil spill software systems in early fate and trajectory of oil spill: comparison analysis of OILMAP and PISCES 2 in Mersin bay spill

The aim of this study is to evaluate the performance level of two advanced oil spill software systems in early transport and fate of oil spill through algorithms accepted in oil spill literature. To do this, the performance level of software systems mostly used in real cases have been compared. OILMAP (the oil spill prediction modeling system) and PISCES 2 (potential incident simulation, control and evaluation system) have been used for spill trajectory in the light of four spill scenarios. The findings reveal that the OILMAP has predicted a relatively larger area of spill. In addition, OILMAP has achieved closer results to the calculations of approaches adopted in the literature for evaporation calculations. Besides, OILMAP software has provided highly reliable results in the evaporation rates of oil compared to the calculations of PISCES 2. On the other hand, as for the determination of the risky area, both software systems have yielded results with high reliability values, which could be used in taking precautions against oil spill in such areas.