of the liner
The first “radar assisted collision”.
At , July 26, the
last of Andrea Doria’s 1,662 passenger and
crew had been evacuated. The primary rescue vessel Ile De France circled
the Andrea Doria one last time, dipping her
colours three times in a farewell salute, then steamed
Inquiries would later determine officers aboard Andrea Doria had used improper radar procedures, and turned to the left at in the moments prior to the collision rather than to the right - the proper Rule of the Road for a head on crossing at at sea.
“The first radar assisted collision”:
She took the northern track against westbound traffic to save time and fuel and with the captain's expressed belief that encountering head-on shipping was safer than crossing. Interestingly, studies done some years later by A.N. Cockcroft and published in the "Journal of the Royal Institute of Navigation" presented data showing that, in restricted visibility, 75 percent of 494 collisions involved meeting situations and 20 percent crossing.
Even if Doria, (thought to be to port) was fine on Stockholm's starboard bow, an early and significant course change to starboard would have complied with the Rules mandating a port-to-port passage in a meeting situation, and more critically, would have supplied early and unequivocal information to Doria's radar regarding Stockholm's intention and thus time to manoeuver if any confusion remained.
Andrea Doria, upon entering markedly reduced visibility, took all the necessary steps (whistle signal every 100 seconds, watertight doors secured, lookout, etc.) but her token speed reduction from 23 to 21.8 knots was insufficient.
Having determined a tight CPA to starboard, she came left (by
only) 4 degrees. The Rules (and courts) mandate a port-to-port passage unless a
starboard passage can be achieved safely without the necessity of either vessel
changing course to open the CPA further. Even this change, if it had been done
earlier and boldly (60 degrees or so), would have made her intentions known to
Doria's excessive speed for the conditions, Stockholm's insistence on going down a "one-way street" in the wrong direction, Doria's attempts to improve the margin of safety for a starboard passage instead of going to starboard for a port-to-port passage as required by the Rules, Stockholm's failure to suspect fog ahead - all of these would have proven moot if the awareness of the developing close-quarters situation had triggered even one of the vessels to "early and significant action."
It's ironic that navigational plots done as part of the subsequent investigation demonstrated that if neither vessel had radar and had thus not changed course, they would have effected a close starboard-to-starboard passage. Radar guided them (and their decisions) into the jaws of collision.
A Destroyer skipper I once had never ceased impressing upon his OODs the idea that it is not only important to know and follow the Rules, but that any action taken be early so that, as he emphasized, "the other vessel knows what you're doing. Make any course change large enough so that (if at night) the light shift can be seen and one that his radar will detect. Half of preventing a collision is you knowing what you're doing -- the other half is that he knows what you're doing." A great skipper.
-- Jim Austin
“Another subsequent radar assisted collision”:
The UK Marine Accident Investigation Branch (MAIB) report on the collision between the British yacht “Wahkuna” and the Liberian flag, containership “P&O Nedlloyd Vespucci” on the 28th May 2003.
This collision has lessons for almost everyone. A large, well equipped, modern yacht, in the hands of a very experienced owner and with a strong crew, was run down and sunk, in fog, in the English Channel by a large, modern, well equipped containership with a good professional crew, which was unaware of the accident and did not stop. Very fortunately, no one was killed, and the colliding ship was successfully identified, so the MAIB have been able to speak to everyone concerned in the incident. This was a classic “radar assisted” collision. Both vessels saw each other on radar in good time. The collision, which would not have happened had radar not been in use on both vessels, was precipitated by mistakes in the use of radar. The ship was keeping a good lookout, with the Master, the officer of the watch and a lookout on the bridge, the yacht had all five crew members in the cockpit and was motoring in a light air with the mainsail hoisted to increase visibility.
The ship was bound from Antwerp to Singapore; at the time of the accident she was making 25 knots, her full sea speed. The yacht was bound from Dielette to the Needles, and was making 7.5 knots under power when she saw the ship’s radar echo. The collision occurred almost midway between Alderney and the Isle of Wight. The French radar station responsible for this area of the Channel, CROSS Jobourg, reported that 19 ships passed through their surveillance area during a period of dense fog, from 10.00 until 14.00, on the day of the accident. Only one of these ships reduced speed. The ship was unaware of the collision. Her officers assumed that no impact had taken place, because the yacht’s echo on the ship’s ARPA appeared to pass clear of the ship by two cables, and the lookout posted on the bridge wing saw nothing.
The yacht’s crew abandoned into their life raft, but the EPIRB did not work, due to a corroded battery. By good fortune, they were picked up, after more than five hours in the raft, by the fast ferry “Condor Express”, which saw a flare. Although the crew of “Wahkuna” did not see the ship’s name, the MAIB had no difficulty in identifying the ship, with the help of the records held by CROSS Jobourg, and MAIB inspectors flew to Hong Kong to meet the ship on her arrival there.
The yacht had a Raymarine radar; besides the chart table display there was a cockpit display and this was being watched. The (M) ARPA features of the radar were not in use and no manual plotting was being done. When the containership’s echo was seen it was assumed that she was on a collision course. The yacht was slowed and the engine taken out of gear, so that the ship could pass ahead. Had the yacht held her course and speed she would have passed eight cables ahead of the ship. The MAIB comment: “The actions taken by the yacht were based on an inaccurate assessment of the situation by radar and served to confuse the bridge team on P&O Nedlloyd Vespucci as well as putting the two vessels on a collision course. Had the skipper been able to make full use of his radar, a more accurate assessment of the situation would have been possible, and it would have been apparent that a substantial alteration of course to starboard, in accordance with Rule 19(d), would have been more appropriate.” “When Wahkuna was virtually stopped in the water, her skipper estimated that the radar contact would pass 1.5 miles ahead as a result of his action. This assessment, however, was based only on a visual interpretation of the radar display, because none of the crew knew how to use the radar’s automatic plotting facilities, and a manual plot was not undertaken… “After making the assessment that the container ship would pass ahead, it is evident that a radar lookout was not maintained. Had it been, it would have been apparent that the container ship was closing rapidly, and avoiding action could have been taken sooner”.
To any yachtsman, the most striking things about the ship are that she was making 25 knots in a fog so dense that her own bow was not visible, and that she did not see the yacht she hit. The MAIB however see things a little differently. They point out that, if the old, pre-Radar, definition of “safe speed” – a speed that allows the ship to stop within the range of visibility – is used, the ship should not have been under way at all, since, even at manoeuvring dead slow ahead she needs three cables to stop. But they proceed to tiptoe rather gently round the whole question of Rule 6 (safe speed) in relation to the containership, with a learned discussion of the Rules, including citations from the leading textbook, Cockcroft and Lamejier, papers by the Seafarers International Research Centre at Cardiff University and the MCA’s own Marine Guidance Notes on Navigation in Fog, and an inconclusive discussion as to whether the Master was under commercial pressure to maintain full speed. This will be very surprising to yachtsmen; it is not at all surprising to professional seamen.
The reality of today is that ships do proceed at full speed in fog, relying on radar, and, in the case of containerships, these speeds are now very high. It will be recalled that, of the 19 ships that passed through the area during the fog, only one reduced speed. The MAIB condemn the ship for having the radar set in use for collision avoidance ground stabilised (referenced to the GPS) and not water track stabilised (referenced to the log and compass). This is contrary to IMO recommendations, because, of course, any tidal stream or current renders the information presented by the ARPA display incorrect. This is an elementary, but common, mistake, so far as radar observers are concerned, and it probably accounts for the ship’s impression that the target representing the yacht passed clear of them by two cables and resumed her course, when in fact she had collided with the ship and was sinking. The MAIB also condemn the ship for excessive reliance on the accuracy of her (erroneous) ARPA plot and for accepting too close a closest point of approach (CPA)
When the yacht slowed down, it was evident to the ship (but not to the yacht, which seems not to have continued to observe the radar) that a risk of collision now existed, but the Master was unsure what action to take; he put the ship in hand steering, posted the OOW to the wheel and posted the lookout to the bridge wing. He did not slow down or alter course, because the yacht’s movements had suddenly become unpredictable to him. The yacht skipper’s decision to slow down to avoid crossing ahead of a large, fast moving target is the decision that most yachtsmen would make. The containership master’s decision to go to hand steering, but to maintain his course and speed, when he became concerned about the yacht’s intentions, is the decision that most professional seamen would take. These two decisions, both quite intuitive, caused the collision, because the yacht’s position vis-à-vis the ship was not what the skipper assumed it to be.
Not surprisingly, the MAIB have written to the makers of the failed EPIRB! They also recommend that the RYA should emphasise the importance of Rule 19 (conduct of vessels in restricted visibility) and that manufacturers of yacht radars should include a voucher for free training in radar plotting with each yacht radar set sold. However, this begs the question of use of radar for collision avoidance aboard yachts.
As this incident demonstrates, unless either manual or automatic plotting is carried out, using accurate inputs, the use of radar in fog can cause a collision that would otherwise be avoided. The yacht in this case was very well equipped, with a cockpit radar screen as well as one at the chart table, and a strong crew of 5. To use radar for collision avoidance requires training, but it also requires that someone who has had that training should keep a continual watch on the radar picture. This is by no means easy to do, and a cockpit repeater is not the place to do it.
With the benefit of 20/20 hindsight, the first lesson that this case teaches is that if a yacht is crossing shipping lanes in fog, and has radar, if at all possible a crew member, trained in the use of radar, should follow big ship practice and maintain a continual watch on the radar set, preferably in the cabin and away from distraction. Rather than apply the old, pre-radar, yacht rule of “When in doubt, in fog, stop the engine and listen” yachtsmen should apply Rule 19 – there is no stand on vessel in fog, think of the trace that you are painting on the other ship’s radar, and avoid turning to port for an echo ahead of the beam.