Oil Spill Trajectory Modelling in Ice

An oil spill trajectory model predicts the movement of spilled oil in and under ice, based on such input parameters as type and quantity of oil spilled and the rate of release, ocean currents, wind speed and direction, and air and sea temperatures.

In Arctic conditions, the predictive task is complicated by the presence of ice in a variety of forms. Models are used for contingency planning where they are useful for decision makers. Modelling a series of the most likely oil spill scenarios allows decisions concerning suitable response measures and strategic locations for stockpiling equipment and materials to be made. Locations shown to be the most vulnerable can be identified and response equipment sited accordingly.

Effective use of models during an actual emergency response requires numerous input parameters. Accuracy and availability of this data can often be an issue. As an incident develops, more precise data will improve the output of the model. In a real-time response application, the models are run to generate forecasts of generally no more than 3 to 5 days duration, recognising that beyond this time scale the wind forecast accuracy degrades quickly. This limitation is no different than how we look at our local weather forecast. Nobody expects a forecast 10 days out to produce reliable results.

Project Reports

Report Sea Ice Model Developments In View Of Oil Spill Forecasting

Recognising that even the best forecast models (oil spill, weather, etc.) will produce ever larger error bounds after days and weeks, it becomes necessary to reinitialise the oil spill models on a frequent basis (tens of hours to days) with the most accurate real-time spill coordinates available, for example using satellite imagery, airborne surveillance data or GPS tracking buoys, and updated wind and ocean current forecasts. Models can indicate whether oil is likely to dissipate naturally or whether it is likely to reach the shoreline. This information can be used by spill responders to decide on the scope of initial aerial surveillance flights and/or the most effective spill response techniques to employ.

The overall aims of this research component were to improve the ability of contingency planners and responders to predict the movement of oiled ice with greater accuracy over longer time periods in a range of ice conditions, including pack ice and the more dynamic conditions of the Marginal Ice Zone (MIZ). The research objectives were: 1) to first improve the resolution offered by the existing ice models, and 2) to provide the ice model data outputs in formats that the existing oil spill fate and behaviour models could efficiently import and process.

The JIP developed a two-phase research programme: the first to focus on improvements to the resolution and accuracy of regional ice modelling, and the second to focus on integrating the outcomes of the first phase into existing oil spill trajectory and fate and behaviour models and to validate or quantify the potential improvements.

This JIP research programme advanced oil spill trajectory modelling by supporting the development of several improved higher-resolution ice drift models that outperform existing models both in pack ice environments with high ice concentrations and more dispersed dynamic ice associated with Marginal Ice Zones (MIZ).

By requiring that the ice model outputs be provided in internationally accepted data exchange formats, the two most commonly used oil fate and behaviour models can now efficiently import the data produced by the best available ice model in any given region to provide more accurate predictions of oiled ice movements in a range of ice conditions The oil spill modelling community and responders now have greater flexibility in the variety of models they can use as well as the ability to rapidly take advantage of new modelling technologies as they evolve.