Case Studies/Oil Spill
English version

The Baltic Carrier accident



The accident

The sugar freighter “ Tern” and the tanker “Baltic Carrier” collided March 29th 2001 at 1:30 MESZ (Danish summer time) in the Baltic Sea at the position 54º 43´ 36´´ N and 12º 30´ 12 ´´E (in Kadetrenden - east of Falster). The Baltic Carrier tanker was carrying 30.000 tonnes of Heavy Fuel Oil 380. The fuel oil immediately began to leak from a hole in the starboard side of the tanker Baltic Carrier. The leaking tank contained approximately 2.700 tonnes of oil, most of which was quickly lost to the environment. Due to the rather rough sea condition (~2,5m waves) and the extent of the boat damages, attempts to control the release of oil all failed.

Oil slick drift observation

The oil spill drifted from the accident location northwest towards the Danish coastline, and reached the Grønsund Strait 17:00 – 18:00 MESZ the same day. The oil continued to pollute the coasts of Bogø, Nordfalster and the souternh parts of Møn. The total affected coastline has been estimated to about 30 – 50 km.

Model setup specifications

In the model the oil release was specified by an instantaneous release of 1000 tons oil and the applied oil type was heavy Intermediate Fuel Oil, IFO 450, which is the oil type, most alike IFO 380, which is available in the model and contain a high fraction of residual fuels.

3D model result

The model predicted oil drift and spreading through the Grønsund Strait between Falster and Møn and an oil pollution of the south coast of Møn, at 17:00 MESZ in agreement with observations. The model predicted oil to drift through the strait between Bogø and Møn. In reality a dam is built between Bogø and Møn, which forces the oil to drift past instead of through.

Figure 1 shows the model predictions results for the oil weathering processes changes as a function of time. 16 hours after the oil spill the model predicted 9.6% oil evaporation and 1.5% oil dispersed to deeper water layers – i.e. 88.6% oil was left at sea surface. In agreement with the observations only a small percentage was evaporated and removed naturally. The main part of the spilled oil polluted the coasts of Denmark – in agreement with observations.

Figure 1. Oil fractions for "IFO 450" oil during the simulation.

Figure 2-5 below, shows the simulation results of the Baltic Carrier oil spill. The blue star indicates the oil spill release position. Red colour indicates oil at the surface, light blue colour dispersed oil and dark blue colour oil deposited at either sea bottom or coastlines.

Simulation series

Figure 2. Simulation time: 29.03.01 13:00 MESZ. (12 hrs after the oil spill). The oil spill drifted towards the Grønsund Strait of Denmark. The oil spill is both drifting at the sea surface and in the deeper water layers.

Figure 3. Simulation time: 29.03.01 17:00 MESZ. (16 hrs after the oil spill). The oil slick has reached the mouth of Grønsund Strait.

Figure 4. Simulation time: 29.03.01 19:00 MESZ. (18 hrs after the oil spill). The oil has polluted the south point of Møn.

Figure 5. Simulation time: 29.03.01 21:00 MESZ. (20 hrs after the oil spill). The oil spill drifted through the strait between Bogø and Møn – which is a result of the missing constructed dam in the model topography. As indicated by the dark blue colour oil has stranded both on the sea bottom and coastlines

2D model result

Using the MIKE 21-SA 2D oil drift model the oil slick was predicted to drift northwest with a southern deflection – causing the main oil spill to strand on the coastline of Hesnæs (Falster) approximately 3 km south of the mouth of Grønsund Strait – in disagreement with the observations.

Comparison of 3D model and 2D model

Figure 6 shows the simulation by the 3D Oil Drift and Fate Model and MIKE 21-SA, respectively. The MIKE 21-SA drift simulation (blue line) deviates from the observations of the oil drift and the oil drift simulated by 3D Drift model (red line). The most likely explanation to this deviation between the two models is the differences in current fields of the models. The 3D drift model applies a 3D current field, while the 2D drift model applies a 2D depth-integrated current field.

Figure 6. Simulated oil drift by the “3D drift and fate model” (red signature) and “MIKE21-SA” (blue line), respectively. Result from the 3D drift and fate model is plotted for each 3-hour and oil drift result by MIKE 21-SA is plotted as a mean track each hour.

Conclusion

In the Baltic Carrier case, the 3D oil drift and fate model predicted oil drift and spreading through the Grønsund Strait and oil pollution of the south coast of Møn – both in agreement with observations. The constructed dam between Bogø and Møn is not implemented in the model topography, which caused the model to predict oil drift north of the Bogø Island and through the strait between Bogø and Møn. In reality the oil drift was west of the island of Bogø.

Reference

Christiansen, B. M., 2003: 3D Oil Drift and Fate Forecast at DMI. Technical Report No. 03-36. Danish Meteorological Institute, Denmark.