Bob Couttie’s Maritime Accident Casebook

Like us? Bookmark this article!

Listen to the podcast

 

 The Case Of The Triple Cross [21:55m]: Play Now | Download

PLEASE READ HERE FOR TERMS AND CONDITIONS OF USE

Maersk Dover (Source: Shipspotting)

We’ll call them Steven, Tony and Vince, not their real names but they were real people on the 3,600 tonnes ro-ro passenger ferry Maersk Dover on 17^th October 2006. They were en route 37 miles from Dover to Dunkirk across the Dover Strait, one of the world’s busiest traffic lanes. Even so, it was expected to be a routine passage of one hour and 45 minutes berth to berth, those expectations weren’t quite going to work out that way.

In a sense, to understand what happened we have to go back in time by a little over a year when the Norse Merchant Group which operated the route was taken over by Norfolkline, a company wholly owned by AP Moller Maersk in 2005. From November that year to July 2006 Maersk replaced the old Norse Merchant Vessels with purpose built ships with bigger capacity. The last of those ships was the Maersk Dover which began operations in August 2006. These ships were so successful that Dunkirk had a record throughput of 56.65 million tonnes of cargo in 2006 and more was expected in 2007.

Everybody should have been happy, and they were, except for Maersk Maritime Services, MMS, which had taken over the ship management and personnel recruitment duties. Its problem was one that we’re very familiar with - a manning shortage. It didn’t have enough experienced officers to pick and choose who it promoted. There was a shortage of UK officers. Recruitment from competitors was difficult because they offered terms and conditions that encouraged their officers to stay where they were. Natural attrition reduced the corporate experience of ferry operations. There was increase turnover of MMS personnel which put a significant burden on training and wasted the benefits of more junior officers understudying their superiors and being mentored by them.

Officers were employed in the highest positions their qualifications allowed. That meant that there was a greater need for monitoring and supervision. It’s a common problem throughout the maritime industry.

Steve, the duty master at the time of the incident, was one the of the Norsk Merchant group officers who had been taken into Norfolk lines. With 23 years experience on ferries, five of them on Dover Strait ferry operations, he was one of the company’s most experienced officers. He’d been aboard the Maersk Dover during her delivery voyage from Korea in the middle of 2006.

Tony was the second officer and new to the company although he’d been a Maersk deck-cadet until he left the company in 2001. He spent five years on ferries crossing the Irish Sea and joined Norfolk Lines in May 2006. In July, 2006, he boarded the Maersk Dover in Port Suez, Egypt, during her delivery voyage on a 25 day appointment. He joined the ship again on 15 October 2006, two days before the incident, for his third two-week tour of duty.

One of the differences Tony noticed between the Irish Sea runs and Dover-Dunkirk operations was just how challenging conditions were on what’s said to be the world’s busiest waterway. By October, it had just become routine.

Vince became a quartermaster in June 2006. The previous year he’d qualified as ‘a rating forming part of a navigational watch’. His duties were to assist the officer of the watch, be helmsman, and keep the bridge clean.

It’s now 6.55 on the morning of October 17, 2006, and the Maersk Dover is berthed at Eastern Dock 3 in Dover Harbour. Aboard her are 64 crew, 135 passengers, 785 freight vehicles and 26 car. The winds are light, the sea state slight, and visibility an estimated five miles.

Vince is cleaning the bridge when the ship goes to harbour stations. He goes to the starboard ARPA radar displays and begins acquiring contacts along the south-west lane of the Dover traffic separation scheme, the TSS. It was a standard procedure that gave the bridge team an early heads up of the traffic situation. There were standing orders to use only the starboard radar for navigation because it was the one hooked into the vessel voyage data recorder or VDR, the maritime equivalent of the ‘black box’ used in aviation.

Steve is also on the bridge together with the chief officer, who is responsible for berthing communications. Tony is in charge of the aft mooring deck.

Steve manouevers the ferry away from the dock with the thrusters and at 7 am the chief mate orders the fore and aft mooring stations to stand down. Tony is due to take over from Steve as officer of the watch so he makes his way forward through the main car deck and the accommodation towards the bridge. It’s about a five minute walk.

Once clear of the dock, Steve turns off the thrusters and transfers manual steering from the bridge port wing to Vince at the centreline console. The true course is set to 095 to pass through the eastern entrance of the Dover breakwater.

At 7.02 Steve tells the engine room that two engines will be needed for the crossing and stands down the chief officer who leaves the bridge.

Tony arrives on the bridge at 7.06 and looks at the port side ARPA display to check the shipping situation. As he acquires targets on the south-west lane of the TSS he and Steve chat about industrial action being taken by another ferry company’s employees and what’s happening to Norfolk line’s business in Eastern Europe.

By now, the Maersk Dover was steering 090 at 21.6 knots.

Steve begins the handover to Tony at 7.13, five cables from the South Goodwin buoy and Vince goes back to his cleaning. First, Steve discusses the contacts on the starboard ARPA display, set to the 6 mile range, north up and off-centred to the top left. There are two groups of vessels travelling south west on the TSS and the plan is to pass astern of one group of three vessels and ahead of the second group of two vessels.

Finally, Steve confirms that two steering motors are working, two engines are running, the ship’s heading 095 true on automatic pilot with a 1.4 knot tide running with her all the way to Dunkirk.

It all takes just one minute and Steve hands conduct of the vessel to Tony and leaves the bridge as the vessel is about to make a course alteration to enter the TSS, something that was common practice on the ferry.

Six weeks before, with a different bridge team, the Dover and the Delft had come into a close quarters situation after a master had left the bridge immediately before a critical maneouvre.

Tony monitors the port ARPA radar. He switches between the 6 mile and 12 mile range scales, believing that it had been set to relative motion, centred and North up. He doesn’t see any vessels approaching in the north east lane of the TSS on the radar.

Nor does he see the UK coastline on the port radar at the 12 mile range, but he should have done. He doesn’t notice that it isn’t there. He isn’t aware that some of the vessels he can see through the bridge window aren’t showing up on the port radar. The radar has not been properly tuned.

At 7.16 Tony begins altering course gradually to 125 degrees true by 7.21. Between 7.16 and 7.26 Tony and Vince chat. As they chat they note a large tanker moving along the south-west lane and see another company ferry, the Maersk Delft apparently running late. The atmosphere on the bridge is relaxed and very informal as Vince continues to clean the bridge.

Suddenly, an alarm sounds behind him Tony goes to investigate it. He believes it has something to do with the GMDSS system. As the alarm sounds, a printer starts at the rear of the bridge. He goes to the starboard aft GMDSS/Sat C terminal and accepts the alarm then pulls the message from the printer.

By now, the Maersk Dover is 2.2 nautical miles from the MPC buoy. It is a key point in the passage, to be marked on the voyage checklist, which signifies that the ship would arrive at Dunkirk an hour later.

An alarm sounds on the bridge

Unknown to Tony, another Maersk ship, the containership Maersk Vancouver, and the Apollonia, a VLCC with a deep sea pilot aboard, are five nautical miles off the starboard bow.

Maersk Vancouver (Source: Sietas-Werft)

Apollonia (Source: Shipspotting)

It’s common to think of containerships as moving slower than ferries but that isn’t the case with the Maersk Vancouver, she’s travelling at the same speed as the ferry and the gap between them is closing at around 30 knots, that’s 6 miles every twelve minutes.

The message Tony takes from the printer is from Maersk’s technical branch in Copenhagen about a problem with the ship’s V-Sat equipment. Someone on the ship had contacted the equipment manufacturer directly to get it fixed. Instead, they should have gone through the technical department and the message tells them to make a requisition through the correct channels.

Tony thinks it’s an important message requiring Steve’s immediate attention so he crosses the bridge to the telephone beside the port side radar and ECDIS, sits on the footrest of the bridge chair in front of them, calls Steve and discusses the message for the next six minutes.

Maersk Dover Bridge, GMDSS Console marked in red

He can’t see through the bridge window. He can’t anything significant on the improperly-tuned port side radar, and he’s not looking at the ECDIS display, which has an AIS feed that should have shown the two approaching vessels.

Maersk Vancouver is now 4.2 nautical miles 50 degrees off the starboard bow and the Apollonia is 40 degrees to starboard 4.1 nautical miles off the starboard bow.

It’s 7.28.

As he cleans the bridge, Vince vaguely notices there’s a couple of ships in the North east traffic lane but doesn’t tell Tony about them.

Vince vaguely remembers seeing two ships on the radar

Nobody is minding the ship.

Tony’s conversation with Steve ends at 7.31. He stands up, looks across the port bridge wing, sees that they’ve just passed the MPC Buoy and takes a range and bearing from the port side radar which Vince logs at 7.32 and goes back to his cleaning. Tony makes a 5 degree alteration to starboard.

Tony sees the MPC buoy, but not the approaching ships

Maersk Vancouver and the Apollonia are now 2.5 miles and 2.7 miles respectively from the Maersk Dover.

Now past the MPC Buoy, Tony makes two VHF calls to Dunkirk West port control to tell them his ETA is one hour. No-one responds. Vince confirms that Channel 73 was selected and Tony tries again, this time, at half a minute past 7.33, he gets a reply.

At that same time, Apollonia tries to raise Maersk Dover but neither Tony nor Vince hear it.

Thirty seconds later, Apollonia calls again on Channel 16. This time Tony hears it. There’s a bit of confusion until they settle on Channel 77. The pilot aboard the Apollonia gives Tony his position and warns him “you are passing too close”.

For the first time, Tony looks out towards starboard and, with shock, sees Apollonia, with her black cylindre raised to indicate a vessel of constrained draught. and Maersk Vancouver bearing down on him at less than two miles, with the Vancouver closing at a half mile a minute.

Tony calls the Apollonia to say that he’s altering course to starboard, but the alterations are so small that the pilot on the Apollonia thinks he’d gone to port so orders 10 degrees of starboard helm in preparation for an emergency full turn.

On board the Maersk Vancouver, the master releases that the ferry will just pass clear ahead. He can’t go to port because of the Maersk Dover and can’t go to starboard because he’s overtaking the Apollonia. He gives five short blasts on the ship’s whistle because he doesn’t think the Maersk Dover is taking sufficient action to avoid a close quarters situation.

Tony also releases that he’ll pass just ahead of the Maersk Vancouver, it’s too late to go around her stern. Shaken, twice he asks Vince to take the wheel and manual steering is engaged. He decides to pass ahead of the Maersk Vancouver, then starboard to pass between her and the Apollonia, then to port to go around Apollonia’s stern.

He doesn’t give heading or courses to steer, just relative headings ahead or astern of the other two ships. Vince used his own interpretation of those orders.

About halfway through the maneouvre he realised that the Maersk Vancouver wasn’t showing on his radar.

Maersk Dover passed 5 cables ahead of Maersk Vancouver, a Closest Point of Approach of 3 cables on her starboard bow. Seconds later she closed on the Apollonia to 3 cables on the port bow, 2.5 cables on the port beam and a final CPA of 1 cable astern.

Images from MAIB report, derived from VDR

Animation of the radar screens

radar2.avi

The Maersk Dover was safely in Dunkirk and secured by 8.52. Dover coastguard, alerted earlier by the pilot as the lose quarters situation developed, confirmed that it would be investigating the incident.

Later examination of the port radar showed that it was working, but hadn’t been tuned properly and hadn’t been set up for the best performance.

This particular incident came to mind after three ships were involved in a collision off Japan in early 2008, with the total loss of one ship and, more importantly, loss of life. That investigation is still going on.

In the case of the Maersk Dover, Apollonia and Maersk Vancouver no-one was hurt and the damage was limited to frayed nerves and, perhaps, a couple of careers. All the same, it’s not something you’d want on your record.

Safety is a bit like that game involving a tower built of wooden blocks in which each player takes out one block in turn. As each block is removed, the tower becomes more and more unstable until it finally falls down.

Accidents are a symptom of a system that’s fallen over because too many of the blocks that make up the system are missing.

In this case, the blocks include everything related to safety, from collision regulations to safety management systems, master’s standing orders, attitudes towards the job, how well the bridge team is focussed on its task, whether they know how to use the resources available to them, their training, their experience and a lot more.

Let’s look at those blocks in no particular order of importance, starting with the port radar.

After the earlier close call between the Maersk ferries Dover and Delft several fleet safety alerts were issued by Maersk. Among the instructions was to basically, continuously make sure both main radars are operating correctly. The port radar appeared to be working fine before the ferry left Dover, but no-one on the bridge had acquired the habit of glancing at both radars or noting whether what they could see out of the window was acquired on the radar.

There goes one block.

Standing orders were to use the starboard radar for navigation because it was tied in to the voyage data recorder. Tony didn’t use that radar.

Out goes another block.

Tony hadn’t acquired that instinct that tells you something’s wrong when something should be onscreen but isn’t, like the coastline.

That’s another block.

None of these particular blocks was important by itself. If all the others stayed in place nothing bad would have happened, but they didn’t.

There is a school of thought that believes that a quartermaster, or helmsman, should just shut up and do what he or she is told but Vince had the skills necessary to contribute to safe navigation, as he demonstrated while the Maersk Dover was in port. If he wasn’t being used as a member of the bridge team then the Maersk Dover was a vessel with a single watchkeeper, Tony, moving at 21 knots through one of the world’s busiest waterways.

I’m going to take out two blocks, one because Vince wasn’t used as a member of the watch and another because Tony was effectively a single watchkeeper and around a third of all maritime accidents happen when there’s a single watchkeeper.

And I’ll take out another because Tony forgot just how challenging the Dover-Dunkirk passage was.

Our tower of safety is now missing a good few blocks, it’s beginning to wobble a bit.

The two men chatted informally, but not significantly about navigation. They weren’t passing the right information, they weren’t sharing relevant information, and that’s very important in bridge teamwork. If Tony had shared with Vince what he saw on the radar and through the bridge window, and encouraged Vince to share information - and the relationship was such that Vince wouldn’t have feared to do so - two things could have happened: Tony, by sharing the information, might have become more situationally aware himself and Vince might just have pointed out that ships seen through the window were showing up on the starboard radar but not the port radar.

So, they weren’t focussed on navigation, that’s one block out. We’ll take out another because Tony wasn’t sharing information with Vince. Also, although Vince spotted the two approaching ships he didn’t tell Tony they were there, so there’s another block away.

Our tower is now getting very, very unstable. We have a single watchkeeper using a dodgy radar as his primary means of determining whether or not there’s a risk of collision.

Of course, it would have been a good idea, given Tony’s inexperience on this route, to give him another warm body to keep lookout. So we’ll take out another block for that.

There wasn’t another warm body, so it was Tony’s job to put Vince as lookout and tell him what to watch for. He didn’t, that’s two more blocks away.

It’s not going to take much more for the tower to fall over.

When Tony heard the GMDSS alarm he could have instructed Vince to keep a lookout, he didn’t, so Vince went on cleaning, there’s another block out of the tower.

Tony took the printout to the front of the bridge, which is where he should have been, but he sat down on the foot rest of the portside bridge chair where he couldn’t see anything except the port radar and appears not to have looked at the ECDIS, with its AIS feed that would have shown the other two vessels. So there go another two blocks.

There’s a fundamental issue here, one that’s covered in STCW: “the officer in charge of the navigational watch shall not be assigned or undertake any duties which would interfere with the safe navigation of the ship“. I’m going to take away another two blocks, the first because Tony did undertake a duty that distracted him from safely navigating the vessel and the second because the company didn’t have a system in place to ensure that officers of the watch were not distracted from their primary responsibilities.

For all intents and purposes, this vessel is now not under command.

When Tony finished his chat with Steve he stood up and looked to port because he wanted to know where the vessel was relative to the MPC Buoy, he didn’t scan to see what else was happening. So that’s another block out.

Once he orientated himself with respect to the MPC Buoy he radioed Dunkirk with his ETA. He did not look out of the bridge window, so another block goes out of the window.

Once he realised he was in a close-quarters situation, Tony didn’t give direct, unambiguous helm orders, that’s another block. The ferry’s initial movement to starboard was so small that the pilot on the Apollonia thought it might have gone to port, so there’s another block away.

…And our tower falls over.

Of course, the incident occurred because of inadequate monitoring of a relatively inexperienced officer, lack of situational awareness, poor use of bridge team resources and the fact that no proper lookout was posted. The last, perhaps, counts more than anything else. It was also the product of the manning shortage that puts less and less experienced officers in ever higher positions on more and more bridges.

That said, it’s important to realise that accidents are rarely the result of one person’s error. They happen because too many blocks have been taken away from the safety tower, they happen because the system no longer works.

The more blocks you take away, the greater the chance of that tower falling, the greater the opportunity for tragedy to occur.

We can never truly reconstruct someone’s mental processes at the time of an incident like this, even with the help of a voyage data recorder, but it would appear that one element was a version of confirmation bias: one makes an assumption then looks for evidence that the assumption is correct. Tony thought the ship was in a safe position, he confirmed that by looking at the portside radar. His Irish Sea experience, on far less busy routes, may have reinforced his assumption.

Had he assumed, despite the lack of targets on the radar, that there was a threat to his vessel he might have looked for it a bit harder: The lack of targets would have raised his suspicions rather than lowered his guard, given the expected high levels of traffic.

Your job is to think continually like the mad German dentist in the movie Marathon Man: “IS it safe? Is it safe? Is it safe?” and don’t take ‘yes’ for an answer.

Follow the rules, follow the safety management system procedures, take notice of safety alerts, question your assumptions and more of those blocks will stay in place and the safer will be you, your ship, and your career.

This is Bob Couttie wishing you safe sailing.

Official MAIB Report