Major Oil Spill in the Mediterranean

French and Italian maritime authorities have commenced cleaning a major spill that has spread 20 kilometers in the Mediterranean Sea after two cargo ships collided north of the island of Corsica on Monday 8 October 2018.

European Space Imaging captured a series of satellite images of the site on Tuesday 9 October at 50 cm resolution with WorldView-2.

The vessels involved in the accident are the Ro-Ro Ulysse, owned and operated by the Tunisian shipping company Cotunav, which ran into the boxship CSL Virginia, owned by Cyprus Sea Lines while it was anchored about 30 km off the northern tip of the island causing upwards of 600 tonnes of fuel to be leaked.

Whilst no one was injured during the collision, the extent of the damage is considerable. Although it is still too early to determine exactly what happened, satellite imagery analysed from before and after the incident can provide key insights to assist the investigation process.

“Satellite Imagery is useful for first responders of any crisis to assess the scope of damage and get another view of the scene. It can also help to monitor and detect any changes that may be critical to minimising or even eliminating further catastrophe” said Adrian Zevenbergen, Managing Director of European Space Imaging. “The benefit of satellite imagery is that the different spectral bands can be used to classify unidentified materials making it an invaluable source of information for maritime disaster management.”

In an effort to contain the disaster, tugboats have been dispatched from three Italian ports and the French navy has also provided a vessel specialised in containing and cleaning up spills with oil spill experts dragging a floating barrier to contain the oil followed by the use of a skimmer to suck up the fuel.

“Satellite remote sensing plays a very strategic role in maritime surveillance and supports the entire response process, from providing initial detection and assessment for situational awareness to directing clean-up efforts” said Dr. Melanie Rankl. “Very high resolution satellite imagery is an additional asset to that of traditionally used radar-based oil spill detection technology. Its ability to provide a detailed overview of the area plays a complimentary role in conducting on-going assessments and monitoring the extent of damage.”

Satellite image at 50 cm resolution showing the collided ships and subsequent oil trail off the coast of Corsica, Italy. The high quality of the data makes it possible to zoom in on the image and thereby identify the extent of the damage and surrounding clean up efforts. ? 09/10/2018 by WorldView-2 © European Space Imaging

Share on facebook
Share on twitter
Share on linkedin

Related Stories

Scroll to Top

Architecture of ResNet34-UNet model

UNet architecture for semantic segmentation with ResNet34 as encoder or feature extraction part. ResNet34 is used as an encoder or feature extractor in the contracting path and the corresponding symmetric expanding path predicts the dense segmentation output.

Architecture of VGG16-UNet model

UNet architecture for semantic segmentation with VGG16 as the encoder or feature extractor. VGG16 is used as an encoder or feature extractor in the contracting path and the corresponding symmetric expanding path predicts the dense segmentation output.

Architecture of ResNet34-FCN model

In this model, ResNet34 is used for feature extraction and the FCN operation remains as is. The feature of ResNet architecture is exploited where just like VGG, as the number of filters double, the feature map size gets halved. This gives a similarity to VGG and ResNet architecture while supporting deeper architecture and addressing the issue of vanishing gradients while also being faster. The fully connected layer at the output of ResNet34 is not used and instead converted to fully convolutional layer by means of 1×1 convolution.

Architecture of VGG16-FCN model

In this model, VGG16 is used for feature extraction which also performs the function of an encoder. The fully connected layer of the VGG16 is not used and instead converted to fully convolutional layer by means of 1×1 convolution.

X