The Helge Ingstad accident report is finally out, or at least the first part of it:
Summary page with visual media
Key factors identified are inexperienced officers on the bridge, piss poor watch keeping and non-existent BRM.
The report comes with video of the impact, recorded from the Sola TS:
In related news, the police have given the captain of the Sola TS formal status as a suspect:
Based on a firmly lodged situational awareness that the ‘object’ was stationary and that the passage was under control, little use was made of the radar and AIS to monitor the fairway.
“It ain’t the things you don’t know that kill you. It’s the things you do know that ain’t so.” – probably Heinlein.
Was the only radar watch on the bridge? In my experience naval ships use a separate CIC which has no windows. A CIC was not mentioned in the report that I saw.
Twain I think.
Fair point, and I would wholeheartedly concur if we were talking about a yachtie single handing on too little sleep, but we’re not. The only way the confusing view out the window makes sense as a main causal factor, is when you’re already headed down the Hjeltefjord doing 17+ knots at night with traffic up ahead, eyeballing the collision avoidance without giving it too much thought. For our nation’s “finest” to do so in this way may be possible to explain, but utterly impossible to excuse.
There has been much speculation about why they didn’t use AIS and radar on the HI, with myself and others floating theories that AIS use may have been off the board for training purposes, and the radar may be switched off to simulate navigation in a non-emissive stealth mode. Now it turns out that they did have both AIS and radar data up on the screens, but were too busy shooting bearings to give it any attention, and weren’t really looking too hard out the window either. The helmsman did, and deftly realized that he was looking at a moving vessel, but figured boys in the back knew what they were doing, and didn’t bother to bring it up. Inexcusable.
And where was VTS? They just “didn’t do like they normally do”?! I pay for that shit! In-fucking-excusable, all of it.
It is hard to see how the manning of the tanker and the lighting of the forecastle can be altered to comply with the investigation. Lights shining aft would distroy the night vision on the bridge and they would not have seen the frigate. The lights are placed high on the foremast to avoid damage, the reason why can be answered by anyone that has experienced heavy weather in a loaded tanker.
It is normal for the master and only one officer plus the pilot to be on the bridge when the vessel is unmooring. The tasks that the inquiry says that the bridge team should have undertaken is unmanageable, remembering that the topping off operations would require the chief mate and another officer to conduct the operation and another officer to test the bridge gear before departure. The master would of been conscious of the requirement to have watch keepers fit for duties in heavy traffic. He is not in a position to breakout a batch of fresh watch keeping officers to watch keep on radar, signal with aldis, monitor wheel orders, keep the log book, shepherd seeing eye dogs on the forecastle, and sound the whistle from some bloody storeroom.
Normally the VTS briefs the pilot on the traffic information and during the pilotage the pilot passes on the information to the master.
If anyone apart from the bridge team on the frigate should be censured it is the personnel in the VTS.
Three on-coming vessel to port, brightly lit oil terminal to stbd. If that was what the OOW was told and visually the picture matched that description exactly, there was no reason to be confused.
An experienced and well trained mariner is more likely to have an established routine to cross-check the visual information but according to the report that was not the case with those officers.
In both cases the behavior would be the same.
From the report:
Another factor relating to awareness at the individual level is that, even when we have surplus
cognitive capacity, we continue to employ economising mechanisms as long as possible,
I had this happen to me, a lot like the Stora – but smaller, of course. I was trudging up the Massachusetts coast under power, no sails up. It was before I had radar. Night time, fisherman with his deck lights on a point or two forward of the port beam going away. I picked him out with binocs
against the lights on shore. Nobody else around.
He was way the hell too close when I realized his working deck was forward and he was about to run me down. If I’d been in a frigate instead of a thirty foot sailboat it would have been all over. As it was, helm over hard and a change of underwear sufficed. I had plenty of capacity but I’d dismissed him several minutes before. He went from irrelevant to imminent danger in (what seemed) the blink of an eye.
I reckon he wasn’t looking out the window, or just didn’t care – but that’s beside the point.
Back in the day the CIC was manned with what we called a cruising watch by two able rate radar plotters under the direction of a leading radar plot rating ( one rate below a junior non commissioned officer).
The senior rating was there on merit and the system was maintained at all times the vessel was underway and if the vessel moved to a higher state of readiness more senior people were employed until at action stations the commanding officer took command from the CIC.
Irrespective on what the bridge were doing, the fact that that a contact had detached from the return of the terminal area and was now a contact underway would have been apparent in the CIC and reported; and that being in the radars I used last century, when valves and mechanical linkages were still flavour of the month.
quotes from summary.
the starboard lookout position was unmanned
The watch changes between the
officers of the watch and the officer of the watch assistants, the night meal and the rotation of
positions between the bridge crew team coincided with the training in optical positioning.
The bridge team was not correctly put together with regards to the requirements for vision in
current regulations. It may be questioned whether the Navy’s system for medical selection
and follow-up was satisfactory.
does this mean the watchouts were were blind?
1.14.2.2 Findings in the vision tests
One of the bridge team members should, … be
assessed as fit for service in the Armed Forces but not for service in the field or at
sea.
One bridge team member was, according to the specialist report, unfit for service in
the Armed Forces under the regulations, including for sea duty and bridge duty,
unfit for service on Norwegian vessels for work requiring a certificate. The
assessment in the specialist report does not correspond to the medical information
from the Navy. The military doctor considered the person in question as fit for sea
duty, but unfit for bridge duty. According to the Department of Occupational
Medicine, the military doctor gave the wrong diagnosis of the condition. There are
deviating findings in the examination, and the regulations have not been complied.
Two of the bridge team members had, according to the specialist report, reduced to
low contrast sensitivity 38 , especially in twilight conditions and twilight with glare.
According to the specialist report both persons were fit for service in the Armed
Forces under the regulations, including for sea duty and bridge duty. The current
regulations do not contain mechanisms for identifying personnel with low contrast
sensitivity in the absence of predisposing factors. The Navy was not aware of the
person having reduced contrast sensitivity.
That interpretation doesn’t work for me. They were already aware that a part of the oil terminal was anomalous, and even noticed that the aspect was changing. If the object was indeed stationary, this would have indicated that their initial position estimate was way off, which should have set all kinds of alarm bells ringing, prompting another couple of bearing fixes or just a quick glance at the radar screen. Instead, the OOW didn’t think too hard about it:
The OOW on HNoMS Helge Ingstad eventually noticed that the ‘object’ on the starboard side seemed to be closer to the frigate’s course line than first assumed, leaving less distance to the closest point of approach. The OOW has stated that the ‘object’ was primarily observed visually and that the OOW did not check the radar for details.
I don’t know about that. I’d expect our yachtie to navigate primarily by visual means, and be fatigued enough to fail to analyze the situation diligently. I have a whole different set of expectations for the crew of one of our warships, or at least I used to. The OOW’s defined duty was to monitor the situation by all available means and intervene if it got out of hand. It doesn’t look like he even tried.
Now you’re touching on logical omniscience as it pertains to navigation, and interesting subject that I wanted to make a thread about. In the end I couldn’t put together a post that stood on its own legs, and deleted the draft, but it may fit here. One of my favorite AI pundits did this little presentation on how having limited information and time to analyze a subject changes our perception. The first half is relevant here, the second half where he gets into mitigation schemes based on financial models doesn’t really translate to the subject at hand:
In this context, the navigator is an agent that seeks to deduce his position with the highest possible precision, up to a level where further precision is not useful, which doesn’t really happen in practice. A navigator performing ded reckoning during passage is logically omniscient; He has all the time in the world to make his calculations, examine the data, cross-check and refine the calculation. Still, it gets to a point where he decides “that’ll do”, because he’s pushing diminishing returns. Herein lies an important difference between computers and humans, but I digress.
Inshore night navigation by visual means is a good example of a task in which you’re far from logically omniscient, but rather bound by observation and processing capacity. In fact, the cognitive load is such that I strongly prefer to have a second person in the wheel house for this task, even in situations of solid margins (as opposed to barreling down the Hjeltefjord at 17 knots). Doing it safely requires a constant regimen of cross checking every fix in sequence, both for positioning the vessel and estimating the movement of others, and there is considerable learned skill involved, beyond understanding the theory and executing it on paper.
The relative difficulty of visual inshore navigation could explain some aspects of this accident, such as how they kept thinking of the Sola as a stationary object in the face of obvious evidence to the contrary. This only works if we assume that the OOW forgot what he was supposed to be doing and got deeply involved in the training task. Even then, this is such an epic, multi layered fuckup that I struggle to bend my head around it.
We might be talking about different things.
It’s possible that the bridge crew on the HI that night were a bunch of worthless lazy clowns who couldn’t have been trusted to ride a merry-go-round. But with regards to the question is how could a person on the bridge see the three ships pass port to port but not see the tanker. The answer is in part inattention blindness, we see what we expect to see.
Maybe the entire Norwegian Navy is deeply incompetent and should be fired. Understanding human factors is still useful.
I’m 100% with you on that one. One pertinent question which may bear some attention due to its universal relevance, is why the people running visual nav on the HI (the OOWT and OOWAT) failed to identify the Sola as a moving object. Seeing as the helmsman figured it out without really trying, I think the answer lies in both attention channeling and confirmation bias.
Another, much more puzzling question is how the OOW and OOWA were distracted from their task of ensuring the safety of the ship by all available means, for long enough that this could happen.
I have had a nasty experience of cognative failure which could have turned nasty and may be relatd to the failure of the HI bridge crew.
We had set off from Dover heading for Cowes in the Solent (UK). It was a 12 hour passage in late autumn. We made the Looe Channel about sunset and all was fine then set course for the forts that guard the eastern end of the Solent.
Sometime after it got dark the GPS failed and stopped updating our position on the ECDIS. I didn’t notice the failure for some about an hour. I was tired and not expecting anything to change so when the radar overlay began to diverge from the chart position… Being tired, I assumed the radar was at fault! (eek).
I was looking all round but failing to understand what I was seeing. It was only seeing the shape of the Horse Sand Fort occluding the street lights behind that suddenly bought everything back to reality.
A lesson learned
This is not at all a surprising result, expected in fact. The OOWT and OOWA are busy with the task of collecting then plotting data. This is a step-by-step process that takes time and attention and only the final result, the fix is usefull. The numbers (the bearings) by themselves tell you very little. Similar to copying the lat and long off a GPS, of no use till plotted.
By contrast the helmsman is staying in one spot on the bridge watching the changes in the relative bearings.
This can be observed in the wheelhouse. Experienced navigators will plant themselves next to the centerline compass with an occasional trip to the radar while the inexperienced will make a constant loop from radar to chart.
I laid this out in detail on this post Ditch the Map. It's Ruining Your Brain - A General Overview of Human Factors and Navigation
Hence my position that this is not the most interesting question at hand. The most relatable, perhaps, but far from the most puzzling. You shared a yarn related to misinterpreting lights in this thread, I shared one (or at least I suggested its existence) in this post, and there are several more in this thread. I think everyone who stood a night watch will be able to relate.
Where it does get a bit weird is with the OOW noticing the changing relative bearing without looking further into it. If nothing else, it underlines that he’d completely forgotten what he was supposed to do.
My outrage, driven by a considerable sense of shame, is directed squarely at the OOW and the VTS operator, who dropped their respective balls and watched them roll away with a shrug.
Well now. Back in the pre-ECDIS days I made good use of known safe bearings. For example, I’d make a mental note of the bearing to a known reference point (say a light) that put me clear of whatever rocks, and kept checking until I saw the magic number, whereupon I could make my turn without worry. I guess this stops making sense once you follow a formal passage plan, and in any case it is a digression best left for a thread on the finer points of visual navigation.
Safe bearings worked for me as well. And when I had a Loran receiver with rudimentary plotting ability on a nice big screen but no charts, I used to determine and drop waypoints on or on the safe side of charted dangers, of which there are quite a few in Maine. I had them marked with some symbol that meant “Don’t Go There”. I found it extremely helpful in the cockpit.
The plotting receiver crashed a lot (I was legendary at the company office for submitting dozens of detailed bug/crash reports, a number of which they fixed) but it was introduced at just the wrong time, shortly before GPS became useful for low-speed precise work; so I’m pretty sure few people bought it and they never finished cleaning up the software. But it was easy to operate, and crashes and all it was a huge help; and I ran it in tandem with another earlier receiver/antenna that worked fine though it had a horrendous user interface and even worse manual. I completely rewrote the manual so other people could make sense of the beast. Both receivers by Apelco incidentally, several years apart – literally the first and last ones they made.
Later I had a GPS handheld with a navigation marks database, basic outline charts, and sharp though tiny display which I used to feed a charting program on a laptop. That gave me three layers of fallback/crosscheck on two navigation systems and no single points of failure including dismasting and losing the electrical system completely, or either the Loran or GPS system itself going temporarily t/u (both of which happened during my cruising time).
Given that a good percentage of my cruising was in comparatively close quarters in fifty foot visibility I found that very comforting and well worth the extra battery drain from running three receivers. And toward the end I had a small JRC radar set that also gave comfort and ease. All the above also let me sail in the fog rather than always motoring since I didn’t have to depend on maintaining a constant speed for the stopwatch, and gauging the current on every passing lobster pot buoy.
Passage plans should be treated as general guidelines but it seems as if some individuals treat them as if they were carved in stone. As the saying goes, it’s a poor craftsman who blames his tools but it’s also a poor craftsman who only uses 1 tool when more are available to do the job effectively. Some people can’t seem to resist the warm fuzzy feeling that complacency provides.
