A cruise liner with 1,500 souls aboard, the most dangerous waters around the US coast and a GPS that tell lies to the autopilot for 36 hours with nobody noticing. Guess what happens next…
Listen To The Podcast
The Bridge Team
Their real names aren’t important so we’ll call the Master Tom, The Chief Officer Dick, The Navigator Joe and the Second officer Harry.
Tom had Master’s certificate and license from Greece and Panama. He’d never been on an integrated bridge before joining the Royal Majesty as Master in 1992.
Dick also had a Master’s certificate from Greece and a Chief Officer’s license from Panama. He’d never been on an integrated bridge before joining the Royal Majesty on `1992.
Joe had a second officer’s certificate from Greece and a second officer’s license from Panama. He’d never been on an integrated bridge before joining the Royal Majesty as navigator in 1994.
By now you don’t have to be psychic to know that Harry had a chief officer’ certificate from Greece, ditto a licence from Panama, and hadn’t been on an integrated bridge until he joined the Royal Majesty in March 1995, six weeks before the incident.
The Royal Majesty
The Royal Majesty was a Panamanian registered cruise ship, 568 feet long and 91 feet wide, built in Finland in 1992 and carrying more than 1,500 souls. She was built with an integrated bridge, a fairly new innovation at the time, which brought all the ship's navigation and control systems onto a single console, including the autopilot.
Well, almost all. The GPS was banished to the chartroom along with the Loran-C display. This was not a good idea.
For the technical guff, you’ll have to read the official report but basically if the GPS receiver lost satellite contact, it switched to dead reckoning. GPS data uses a standard code called NMEA, designed by the National Maritime Electronics Association. When it was put together nobody thought there’d be any sense in a GPS receiver sending out dead reckoning data, and really there isn’t, so there wasn’t a code for it. The GPS designers found a way around the problem, it sent the data with a bit set to ‘invalid’, a sort of note to say ‘ignore me’.
GPS data went to the autopilot. The designers of the autopilot thought that if the GPS receiver didn’t get satellite data, it would simply stop sending data. They were wrong, so they didn’t program the autopilot to look for the note that said ‘ignore me’.
When the ship was delivered, she had a GPS that guessed where it was when it lost satellite contact and an autopilot that believed everything the GSP said.
The ship had radars either side of the bridge equipped with automatic radar plotting aid, an ARPA, which overlaid an electronic map image supplied by the autopilot.
When the GPS lost satellite signal, it sounded a rather feeble alarm, the sort of thing you might have on a wristwatch, for just one second. An external alarm could have been fitted, but wasn’t.
Because of a problem with satellite reception the GPS antenna had been moved from its original position and the cable ran across the roof of the bridge and wasn’t fixed where it could not be accidentally kicked free.
One other bit of equipment is worth noting, the echo sounder, referred to in the official report as a fathomometer. Normally, this was set to sound an alarm if the depth under the keel was less than three metres, except in port, when it was adjusted to zero metres.
Royal Majesty chartroom with GPS. A second Loran-C display has been added since the incident.
At 1200 on June 9, 1995, The Royal Majesty left the dock at St. Georges, Bermuda, on the 667 mile return leg of a cruise to Boston, Massachusetts with good weather and clear visibility. Joe checked the GPS and Loran-C equipment, it was working.
Nobody had reset the echo-sounder alert to three metres from its in-port setting of zero.
About 52 minutes after taking to sea the GPS antenna was somehow disconnected, possibly kicked loose by a crewman. The GPS receiver defaulted to dead reckoning. No-one heard the alarm.
The Royal Majesty GPS display. Notwe SOL - Satellite On-Line indication
For the next 24 hours the vessel followed its planned course of 336 degree at a little over 14 knots. Hourly plots were out on the paper chart based on the GPS readout – no-one noticed that the GPS was on dead reckoning. The Loran-C was not checked because it was only regarded as a back-up, not as an alternative means of checking position.
On the evening of June 10, chief officer Dick compared the GPS and Loran-C readouts twice and they seemed to agree.
Several times, Master Tom phoned the bridge asking about the buoy that marked the southern entrance to the Boston traffic lanes, known as the BA buoy. Dick expected to see the buoy at around 1845. At that time, he saw a radar target which, based on the GPS reading, he took to be the BA bouy. He did not visually confirm the target because it was hidden in the glare of the setting sun.
In fact, it was an entirely different buoy, known as the AR buoy which marked a wreck. But it was in the position that the chief officer expected for the BA bouy.
BA bouy (above centre) and AR Bouy (left of centre). Heavy dashed line
to the right is the assumed track by dead reckoning. The lleft heavy
dashed line is the actual track. Note Phelps Bank etc. dead ahead on
Harry took the watch at 20.00. Half an hour later a lookout reported a yellow light to port and a few moments later both the port and starboard lookouts reported high red lights. Harry took no action. These lights were not mentioned to the master when he came to the bridge for the first time during that watch.
The Royal Majesty was not alone on the sea. Two fishing boats with Portuguese-speaking crews saw her. They tried to call her in English on Channel 16 but got no reply. The two boats discussed the out-of-place cruise liner
Tom was on the bridge again at 2200. He didn’t check the vessel’s position because the second officer had told him they were already past the second bouy of the Boston traffic lane, bouy BB. In fact, he hadn’t seen the BB bouy, either visually or on radar but based his assumption on the GPS reading.
The vessel was already moving over shoal that would have triggered the depth alarm if the fathomometer had been correctly set.
Shortly after Tom left the bridge at 22.10 the port lookout reported blue and white water dead ahead. Harry did not respond.
Suddenly, at 22.20, the ship veered to port. Harry switched the helm to manual steering. The movement of the ship brought Tom to the bridge. The starboard radar showed the vessel to be less than 10 mile from Nantucket. He checked the position in the chartroom and immediately ordered hard right rudder but it was too late. Before the helmsman could respond, the Royal Majesty ran out of water on the Rose and Crown Shoal, 17 mile off course.
The Second Officer told the Master the Royal Majesty had passed the BB Bouy (to right, on dead reckoning track)
but in fact was already over shoals.
Instruments of Disaster
Navigational aids are just that, aids. They don’t navigate, people do. True, had the GPS been working properly, or the fathomometer been correctly set, the Royal Majesty probably would not have grounded, but it was the reliance by the ship’s officers on these instruments that led to the grounding.
There were only two checks in those hours before the grounding on the accuracy of the GPS, comparing it to the Loran-C reading, which appeared to agree within the necessary limits so it seemed that the GPS was correct and working. Yet, every time an officer took a reading from the GPS to plot on the paper chart, staring at him from the screen, just below the ship’s position were the letters ‘DR’, for dead reckoning. Nobody noticed them.
What we’re dealing with here is confirmation bias. We tend to look for data that confirms our expectations, if it fits, we don’t question the data. Because the GPS and Loran-C seemed to agree, nobody noticed that the GPS satellites were offline and vessel was on dead reckoning.
That wasn’t the only example of confirmation bias. The Chief Officer saw a radar target at the time and place he expected to see the BA bouy but could not get visual confirmation. All the same, he informed the master that they had passed the BA buoy.
Data that seemed to confirm their assumptions was accepted, data that did not, like the high red lights reported by the lookout, was discarded.
In a sense, the master was rather like the autopilot, he was making assumptions from partial and incorrect data, data that appeared to confirm his assumptions about the ship’s position. When Harry told him they had passed the BB bouy the master did not know that the bouy had not actually been seen visually or on radar. Because of that, Tom did not check the ship’s position – he thought he knew where it was.
Those assumptions resulted in the vessel being 17 miles off-course in some of the most dangerous waters around the United States coastline.
If Harry had told Tom that he hadn’t seen the BB bouy, Tom would have made more efforts to determine the ship’s position.
It takes just a split second to check whether the GPS shows SOL – satellite on line or DR – dead reckoning and it’s probably a good habit to get into. It’s a better habit to regularly use an alternative means of position fixing.
Man and Nature
Yet man and nature were trying to send a message to the bridge of the Royal Majesty. The lights spotted by the lookouts might have raised Tom’s suspicions, had he known about them but Harry did not mention them. Had the Royal Majesty been on the right course, those lights should not have been there.
The source of the lights did not appear at the 6 mile range on the radar. If Harry had checked at the 12 mile range, or had one of the radars on the 12 mile range he’d have seen Nantucket island and at that range the radar map on the ARPA display would have shown beyond doubt that the vessel was in the wrong position.
The Royal Majesty radar with ARPA radar map.
On the six-mile range the map concurred with the presumed position.
If he’d looked at the chart he’d have seen that the nearest red light to the presumed position was the bouy on the Rose and Crown Shoal.
The Channel 16 radio call from the fisherman was at first ignored, then a voice, believed to be Harry’s, told them to be quiet. There is a requirement to monitor emergency Channel 16 at all times but it’s equally important to listen, because the ship they’re talking about might be yours.
And, of course, a lookout reported blue and white water dead ahead just 10 minutes before the grounding. Blue and white water may look good on the tourism brochures but it’s not normally what you want to see in front of your ship.
It would have been a great time to take a look at the echo sounder display on the console, but Harry didn't.
Why were these signs ignored?
It points to a danger that has proven significant in the air transport industry, too. Technology brings many benefits to the bridge but it threatens to isolate the ship’s officers from the environment they work in. Increasingly, they are being called upon to be systems managers, not seafarers. The attitude is that if the system doesn’t tell you there’s a problem, then there isn’t a problem, despite what you can see outside the bridge window. But what you see out of the window may tell you more than the thousands of dollars worth of electronics programmed by people whose closest connection to the sea is a bottle of bath salts.
Even looking out of the window isn’t much use if you don’t relate what you see to the safety of your ship. After all, an officer of the watch is supposed to watch out for the ship.
Asking the Right Questions
Asking the right questions can go a long way towards getting the right answers: how do I know the GPS is working properly? Is that really the bouy I'm looking for? How has it been verified? Why haven't I seen the bouy I was expecting to see? Is that light supposed to be there? Why is the sea an odd colour? Is that my ship they're talking about on Channel 16? Isn't it time to check position using something other than the GPS? What will I see if I expand the radar range?
Navigation technology will get more complex and invasive. Only two living beings will inhabit the bridge of the future, the master and a dog. The master will be there to feed the dog and the dog will be there to bite the master if he touches any of the equipment.
For now, try not to make a dog's dinner of navigation.
If you have have opinions on this, or any other Maritime Accident Casebook, episode, or want to share your experiences, go to maritimeaccident.wordpress.com or drop me an email at email@example.com
This is Bob Couttie wishing you safe sailing.