The Mauritius Oil Spill Ship Was Sunk At An Undisclosed Location. Where Did It Go?

A huge bulk carrier from the largest category rises from the ocean, tens of meters up into the sky. ‘Hey, rocket to the moon’, says an ironic voice in the background. Cheering when the vessel almost reaches an upright position before it starts sinking away into the depths of the sea.

This is the scene of a video that was released by the Mauritian government to take away concerns that explosives had been used for the sinking of the wreckage on video, the MV Wakashio.

The wreckage is Wakashio’s bow section, which broke off from its stern weeks after the vessel crashed and got stuck on a coral reef just off the Mauritian east coast. The crash and major oil spill left many unanswered questions, but this was also true for the subsequent clean-up, the salvage of the front section and the dozens of whales that washed ashore shortly after.

Gif 1

The oil disaster unfolding between 17-7 and 27-09. The spill is clearly visible until the end of August, and even seems visible September 3. | Imagery courtesy of Sentinel Hub, custom script by Pierre Markuse

Environmentalists opposed the decision to sink the ship. For example, Greenpeace warned that “sinking [the] vessel would risk biodiversity and contaminate the ocean with large quantities of heavy metal toxins, threatening other areas as well, notably the French island of La Réunion”. Authorities addressed these critiques and stated that various NGOs and experts were consulted. Nevertheless, this did not entirely take away the concerns and even the legality of this operation has been questioned. And why didn’t the authorities disclose the location? Would it be possible to find it with open source data?

What Do We Know?

The expedition guiding the front section to its final resting place started on August 19, when two fire fighting vessels with Maltese flags under control of Dutch SMIT Salvage — the Boka Summit and Boka Expedition — began towing the bow eastward towards open sea, leaving the stern behind on the reef. Days later, on August the 24th, the National Crisis Committee confirmed in a press release that the “planned sinking of the stem (forward) section of the casualty has been completed and at around 3.30pm was no longer visible on the sea surface”. The release included three images, but no reference to a location. However, reference to the position was made days prior press release, and would later be made several times.

Image 1

The Boka's dragging the front section off the reef on August 18. | Satellite imagery courtesy of Maxar

A first hint of the actual location was shared in a tweet by what seems to be senior advisor to the Mauritian government, Ken Arian. On the first day of the expedition, Arian shares a latitude, 20°22’ S, and a sea depth. Although the longitude is missing and only degrees and minutes of a latitude are given, the included sea depth makes it possible to bring the area down to a relatively small area. Following the latitude band between 22’ and 23’ from the east coast of Mauritius eastward down the sea bed until 3180 meters, leaves an area meeting the criteria of about 2 nautical miles long, about 14 nautical miles off the east coast.

Image 2

Tweets from government official Ken Arian and journalist Zeenat Hansrod.
The timestamp of Arian’s tweet is local time (MUT).

Other areas also meet the criteria, such as a similar area off the west coast, and several areas more than 90 nautical miles away from the crash site. But we can ignore those, as is shown later. And there’s also another source pointing in the same direction: Mauritian journalist Zeenat Hansrod claims exact coordinates of a location confirmed by the government. The location turns out to be in the same area as the one derived from Arian’s data.

Image 3

The coordinates from journalist Hansrod and the area meeting the requirements given by Ken Arian in white off the Mauritian coastline. The red triangle marks the crash site. | Satellite imagery courtesy of Maxar

In the days following the press release, more statements were made. On August 28, the Minister of Environment mentioned a depth of 3100 metres. Other numbers found in statements range from 11.2 nautical miles to 13 nautical miles from Pointe D’Esny, and a depth of 3000 metres. Even though the differences in numbers are relatively small here, it’s noticeable that the numbers differ per statement, raising the question whether these numbers are accurate.

Verifying The Area

The first of three images (Image 1 from here on) released with the press release reveals a clue. Increasing the contrast of this photo makes a coastline visible on the horizon. The view resembles the view of the Mauritian east coast as seen in Google Earth’s terrain view in western direction from the perspective of Hansrod’s coordinates, roughly confirming a larger area off the coast that includes the positions from Arian and Hansrod. This finding also makes the range of 11-13 nautical miles from the coast plausible. From the Boka’s bridges one should be able to see, roughly speaking, about 10 nautical miles till the horizon, but hills and mountains can be seen from further away. This corresponds with only the contours of the hills being visible. As the bow is visibly making a lot of water, the image should’ve been taken shortly before the recording of the released video, hence close to its final destination.

Image 4

Left: Image 1, released with the press release on August 24. Top right: Slightly altered and magnified horizon.
Bottom right: Terrain view of the Mauritian east coast.| Maxar / Google Earth / Mobilisation Nationale Wakashio

Image 5

Part of the coastline visible on the horizon of **Image 1**. | Satellite imagery courtesy of Maxar

Still, this area is a large haystack, even if the needle is a 300-meter-long vessel stem. Getting hold of part of the automatic identification system (AIS) tracks of the two tugboats that were seen leaving with the Wakashio wreckage, helps to further narrow down the search area.

Mapping The Scene

The tracks of the Boka Expedition and Boka Summit on the 24th of August can be divided into three phases corresponding with the squares shown below. In the early morning, movement took place in area A, an area about 1 square nautical mile, where the tugboats were circling in each other’s proximity. Then, after noon, both tugboats suddenly started moving away from this area towards area B. Until later in the afternoon, when the Boka’s made larger circles in area C at a seemingly quicker pace before sailing back to Mauritius.

Image 6

AIS tracks of the Boka Summit and Boka Expedition on August 24, slightly zoomed in. | AIS tracks courtesy of FleetMon

The video and images paint a pretty clear picture of the scene that could be converted into a two-dimensional image that fits the map. The first indication that stands out is that the Mauritian coastline is not visible on any imagery, except for Image 1. In addition to this, the Boka’s are positioned in opposite direction on the same axis. This suggests the imagery is taken in a north-south direction.

image 7

Left: **Image 2** from press release.
Right: Still from video, at about the same time as **Image 2** was captured.
| Mobilisation Nationale Wakashio and lexpress.mu

The direction in which the noses of the tugboats are pointed, the heading, can be verified by looking at the direction the splashing water is blown by the wind. The wind direction was east-south-east that whole day, so if the tugboats are orientated on a north-south axis, the water should be blown at right angles to the tugboats. It does exactly this. If this is correct, then the video was taken in northern direction and Image 2 was taken in the opposite direction towards south.

Gif 2

Video excerpt that shows the splashing water is blown at right angles to the view. | Video lexpress.mu (Annotation by author)

The next step is the identification of the tugboats. Despite the Boka’s appearing to be identical, Image 2 shows part of one vessel’s stern, exposing a useful and distinctive detail. Going through pictures of both vessels over time shows that the configuration of the orange railing is different for both tugboats. The Summit not only features different cut-outs, it also has two lights installed — which all match the railing visible in Image 2. It’s even possible to verify this by looking at the other vessel, thanks to a piece of the railing entering the frame of the video in a split second around 3:00, revealing a cut-out that matches with the Expeditions’. This means that the tugboat heading north equals the Summit and the video was recorded from the Expedition.

Image 8

Top left: Boka Summit Photo WG7000, FleetMon
Top centre: **Image 2** | Mobilisation Nationale Wakashio
Top right: Boka Summit | Photo SMIT Salvage, LinkedIn
Bottom Left: Boka Expedition | Photo Dolphin Offshore Chandling, Facebook
Bottom right: Video still | lexpress.mu
All images have been cropped.

Now, it should be possible to put the emerging picture on the map. Area A is the area with Hansrod’s coordinates right in the middle. Yet, area B draws more suspicion. There are two reasons for this. To begin with, in area A, the tugboats are largely moving synchronously in each other’s proximity. This doesn’t match the imagery, where the vessels are positioned in opposite direction at a reasonable distance from each other.

Then there is also the shadow of the orange railing of the Summit, visible in Image 2. A blueprint of the vessel can be used to measure the length of the shadow in proportion of the railing — resulting in a shadow length about equal to the height of the railing. According to SunCalc, a 1:1 ratio occurs twice that day: around 10:00 and 14:30. At 10:00 in the morning, the sun’s azimuth (horizontal angle) is such that the tugboat should have been turned so that its heading would be far more east to get the shadow as visible in the image, which likely would have resulted in a visible coastline on the horizon of the image, which is not. In the remaining hours in area A, till around 13:00, the shadow is shorter. This makes the afternoon more likely.

The exact time the shadow equals the height of an object, in the afternoon, is 14:22. A minor correction due to the rolling angle of the vessel — measured by the angle between the deck and the horizon — brings the estimate to 14:37. The tugboats were located in area B around that time.

Image 9

Blueprint of the Boka Summit with the shadow length of the orange railing at different times (MUT) on August 24 as calculated with SunCalc. Important to note is that ship roll motion influences the shadow length.
| Blueprint boskalis.com

The Approximate Location

The released imagery illustrates a picture in which the Wakashio submerges on a location in area B approximately in the middle between the Boka Summit, heading north, and the Boka Expedition, heading south, somewhere around 14:37. This illustration seems to be a good fit with the tracks at around 15:00.

Image 10

Tracks of the Boka’s further zoomed in on area B. The green line is the track of the Boka Summit and the orange track the Boka Expedition. The white arrows indicate the headings of the tugboats at that specific AIS position, the coloured arrows the possible line of sights as seen in the imagery. | AIS tracks courtesy of FleetMon

The get to the approximate location of the Wakashio, a line can be drawn between the positions of the two vessels on the map to determine the middle. Looking at the released imagery in close detail, Summit’s axis almost perfectly aligns with both the Wakashio and the Expedition at the moment Image 2 was taken. This fits best with the position of the Summit at 14:51. Drawing a circle — with a radius of about 250 meters — at a point halfway the line between both vessels, marks the area where the wreckage disappeared on the surface, the centre of this circle being at 20°23’17.22”S,58°0'6.08"E.

Image 11

AIS tracks of the Boka’s zoomed in even further on area B, with a line drawn in yellow indicating the line of sight from the Summit if **Image 2** had been captured at that moment from that position. A circle is drawn in between the tugboats, to mark the area the Wakashio was likely sunk. | AIS tracks courtesy of FleetMon

Discussion & Conclusion

A narrative that fits the picture, is one in which the scuttling started near the intended location in area A, close to the coordinates of Hansrod, before the bow section slowly drifted to area B — accompanied by the tugboats on either side as the tracks show — after releasing the towing cables to let the Wakashio sink. Then, after the sinking at location, the salvage crews made a few more circles in area C before sailing back to Mauritius.

The suspected location is about 13 nautical miles from the Mauritian coastline, at the edge of territorial waters. The sea depth in the circle marking the position ranges from 2675 to 2833 meters with a depth of 2752 meter in the centre.

Image 12

Measured distance from the centre of the circle to the crash site and nearest coastline. | Satellite imagery courtesy of Maxar

Striking about this location is that it is in a zone that has been labelled* as having a strong westward current for a few months a year to the east of neighbouring island La Réunion. It is known that Réunion did share a request with the authorities about the position.

Important to note is that while the AIS track of the Summit unexpectedly included two positions close to 15:00, the printout only disclosed one detailed position per hour, and the tracks are based on no more than four positions per hour. The straight lines between these positions may give a distorted picture, as the actual headings and movement in between the dots is unknown. Having only this data available, makes it impossible to establish with certainty that the suspected position is the only moment that the imagery fits the tracks, as the released video only covers as little as 3 minutes.

The shadows on the orange railing are also up for discussion. Although these seem to support the theory, the measurements are not very precise. Lastly, while the historical tracks of both vessels have been analysed from the moment they left the coral reef until their return to the Mauritian coast, the tracks preceding the 24th were not analysed in great detail.

That all being said, the approximate location looks solid, and independent analysis conducted by analytics firm Windward points to the same area*. There also seems little reason for authorities to misrepresent the day and time. And the narrative that the tugboats would have been circling around on the 24th on a location that is different from the location the Wakashio was sunk in the preceding days seems very unlikely. Despite this, the question of why the Wakashio was initially towed to a location further south remains unanswered.

The larger area the Wakashio sank is verified by the visible coastline on the horizon of Image 1. The smaller area marked by the circle not only matches with the shadows, the direction of the wind, the heading of the vessels and the absence of visible coastline on imagery, but it also fits the AIS tracks and the narrative of the tracks as a whole.

Now that 15:00 is the suspected time the Wakashio submerged, it should also be noted that something very similar to ‘1500’ is heard through radio communications on the video’s audio track just after the wreck disappears from the sea surface.

Audio track cut of the released video from 2:57-3:11. ‘1500’ seems to be heard through radio at 0:06. | lexpress.mu

While we cannot say with certainty that the Wakashio wreckage lies on the bottom of the sea at the suspected position, if anyone asks me to look for the Wakashio on the seafloor, I know where to look first.

* While doing this investigation, a Forbes contributor published an article on the position, which included analysis by maritime analytics firm Windward. It was published on October 23, at a time I had largely completed my investigation. As Forbes’ publication only disclosed a larger area, I independently finished my investigation and blog.

With thanks to Patricia Beekelaar.