Forecasting the Erika oil spills

Abstract

On December 12, 1999, the Erika tanker broke in two sections at about 30 miles from the Brittany coast in the Bay of Biscay, France. The two parts of the wreck sank a few hours after the break. Some 15,000 tons of heavy fuel were released into the marine environment. It is the most serious discharge that has occurred in France since 1980 (Tanio, 6,000 tons). The nature of the incident, the kind and quantity of oil spilled, and the prevailing weather conditions posed considerable response problems. The spilled oil drifted for 2 weeks before reaching the coast. Three different models were implemented by CEntre de Documentation de Recherche et d'Expérimentations sur les pollutions accidentelles des eaux (CEDRE) within a couple of hours of the Erika sinking. On December 14, it appeared that the forecast of the MOTHY model was closer to reality. The MOTHY model was developed by Météo-France (the French national weather service) to simulate the movement of pollutants in three dimensions. MOTHY is an integrated system that includes hydrodynamic coastal ocean modeling and real-time atmospheric forcing from a global model. Pollutants can be oil or floating objects. CEDRE contributes to the improvement and validation of the model using both experiments and interventions during actual pollution events. New developments, exercises, and training are jointly conducted. In the event of marine pollution, Météo-France sends meteorological forecasts and pollutant drift forecasts to CEDRE. This response system has been operational since February 1994. The MOTHY model was used routinely for several weeks after the ship broke up. The model predicted that the coastline was at risk and that the beaching of the main slick would occur after 2 weeks. Diffuse pollution reached the coastline I or 2 days before the main slicks, about 200 km west of the main beaching. Hindcast runs and backward integration of the model explained this unexpected arrival of oil. Some pollution was still arriving onshore several weeks after the initial release. This longer-term pollution came from the wrecks, but also of older pollution by the coastal detachment and deposit tides. Using the model in conjunction with remote sensing information allowed operators to develop and then execute a response strategy rather than react only to observed information.