This post is going to be a little different from previous, as it concerns the recent collision and grounding of the Norwegian frigate HELGE INGSTADT after a collision with the tanker SOLA TS. This is something I take a close interest in, both personally and professionally, and the abundance of open source information allows me to make a few educated observations that some may find interesting.

From the very outset it is important to set several conditions. I have no inside knowledge of what happened, I am working purely from open source information and I want to try and paint a picture of what might have happened that night. The crew are thankfully all safe and it is clear that heroic action was taken to ensure every single crew member was able to disembark quickly. Any analysis in the absence of official information could easily lead to the assignation of blame, but I cannot and will not do so here as we do not have a full picture with all relevant facts. Nevertheless, I believe there is value in providing something educational, that illustrates some of the pressures we place upon our younger watchkeepers and the huge responsibility they shoulder.

The information I have sourced comes from three principle sources. The first is a release of the local Vessel Traffic Service (VTS) radar surveillance data, with radio calls in Norwegian overlaid (link). The second is the same radio calls, without surveillance but with English subtitles (link). A further useful source is a video of the AIS track movements (link). The correlation of these different sources gives a moderately high level of certainty that the tracks displayed are accurate and have not been modified significantly. However, some editing has taken place and the data must still be treated with caution; I will illustrate those cautionary issues as best I can. Other information sources include the International Regulations for the Prevention of Collision at Sea (the IRPCS or ‘Rules of the Road, the maritime manoeuvring rules summarised here), NATO public relations material on Exercise TRIDENT JUNCTURE, and the Norwegian maritime safety service website.

Environment

This took place near Bergen, about the same latitude as the Shetland Isles. The area of water, about 2-3 miles wide, marks one of the northerly approaches to Bergen itself and is controlled by the Fedje Vessel Traffic Service (VTS)(link). A quick search shows the extent of the reporting area (see below); vessels entering this area of water must report in on one of two VHF radio channels and must do so some way before entering the confined waters. An additional observation is that normally in VTS controlled approaches, vessels with serviceable Automated Identification Systems (AIS) would be expected have these systems switched on to enable the VTS to do its job.

The island west of the collision location is Alvoyna in the Oygarden municipality, and the oil terminal at Sture is a busy port that receives large tankers such as the SOLA TS which was involved in the collision. Otherwise the islands in this area are relatively sparsely populated with just 4,900 or so inhabitants across the whole municipality. From that we can assume that there would be few lights at night to mark the shoreline, with the exception of the oil terminal which would be brightly lit. The channel is approximately 2 miles wide but is relatively deep, narrowing south of Sture into a channel little over a mile wide.

The collision occurred at approximately 0400 local time on Thursday 8th November 2018. Sunset the previous evening was at approximately 1623; sunrise would not be until 0821. There was no moon; the moon set at around 1718 the previous day and would not rise until 0847. We cannot be certain of the weather conditions which may have restricted visibility.  However, there is no evidence of weather or high seas on the radar picture and if visibility was restricted by rain or fog, a ship would be unlikely to be sailing at high speed.

Finally the ships involved. SOLA TS is a Maltese flagged oil tanker of 62,000 tonnes. We know she had 23 personnel onboard and after the collision was reported to have little or no damage. Merchant ships are usually well built, especially when carrying petroleum cargoes which if leaked could have devastating environmental consequences; consequently the lack of damage is hardly a surprise. As a result they are sluggish, slow to manoeuvre or accelerate/decelerate. They are not, however, considered to be ‘restricted in their ability to manoeuvre’, a special condition identified in the International Regulations for the Prevention of Collision at Sea – the IRPCS or ‘Rules of the Road’ – this is a condition applied only to vessels which are restricted by their work, such as picking up or laying submarine cables or pipelines, launching or recovering aircraft, carrying out underway replenishment, etc. SOLA TS might have been slow to manoeuvre, but she is not exceptional and is unlikely to have carried any special status. SOLA TS had a tug, TENAX, in company and might conceivably have been considered to be under tow; however, once again no special status is conferred unless the nature of the tow made it particularly difficult to alter course. The tug is more than likely to have been pacing the tanker, probably not connected and likely a precaution for a fully laden oil tanker in narrow waters.

HELGE INGSTADT, by contrast, is a FRIDTJOF NANSEN-class air defence frigate of just 5,290 tonnes. Lightly built for speed and manoeuvrability, warships are invariably less robust than merchant ships, but are more tightly compartmented and have more complex damage control arrangements to compensate.

Narrative

From the video clips above, a rough timeline of the events can be determined. Noting the presence of editing in these tapes, I have avoided trying to extrapolate exact timings. However, relative timing and positional information is still valid.

We know that SOLA TS sailed from Sture Terminal and proceeded north, with the tug TENAX in company. Another tug, AJAX, assisted her departure but returned to Sture immediately afterwards.

SOLA TS was 0.5 miles north of Sture when the first radio message relating to the incident is believed to have been transmitted, a conversation between SOLA TS and Fedje VTS asking about the identity of a radar contact 2 miles north of SOLA TS, and heading towards her; this is the moment depicted above. The SOLA TS had to turn through 180 degrees when she left the berth and her bridge team may not have seen the HELGE INGSTADT until clear of the terminal; however, her radar would have been expected to detect HELGE INGSTADT in good time and Fedje VTS should have advised her of a fast southbound contact.  It is interesting to monitor the third video above (showing only AIS contacts) in which the departure of SOLA TS appears to be delayed until a southbound merchantman is clear.

Fedje VTS initially responds that they do not know the identity of the radar contact and state that the vessel had not reported to them. This is unusual – the Fedje VTS area of operations extends well north of the area and transiting vessels should have checked in. After a pause, Fedje informs SOLA TS that the vessel is probably the HELGE INGSTADT; a few seconds later, contact is established between SOLA TS and HELGE INGSTADT.

An illustration (not definitive) of what HELGE INGSTADT might have seen looks something like this:

At this point in time, SOLA TS is slowly accelerating northwards at 6-7knots, whilst HELGE INGSTADT is just 2 miles north of her, on a reciprocal course, at 17kts. That’s effectively a head-to-head situation with a closing speed of 24+ knots. Now 2 miles at less than 30mph sounds like plenty of time to assess a situation and react. It is worth highlighting that at that rate of closure the ships had about 5 minutes to act (24kts means you travel 2.4 miles every 6 minutes and 0.4 miles every minute). Let us look at the other factors.

East of SOLA TS was a group of merchant vessels following the channel northbound and overtaking her (4 knots faster) – the ships SILVER FIRDA, VESTBRIS and SEIGRUNN (plus a possible fourth contact not transmitting on AIS). From SOLA TS perspective, these ships severely limit her ability to turn away from HELGE INGSTADT. In theory this should not have been a problem – overtaking ships are required by the IRPCS to keep out of the way of ships they are overtaking (Rule 13). They have, in my view, sufficient safe water out to the east to give SOLA TS more room to manoeuvre, in particular SILVER FIRDA and VESTBRIS.  But no request for such room was made over the recorded VHF channel by SOLA TS or HELGE INGSTADT.

Returning to the narrative, SOLA TS called HELGE INGSTADT again and insisted she turn to starboard (in this case, to the west towards the island of Alvoyna. HELGE INGSTADT replied that she could not do so, as to alter westwards would put her too close to the shore (‘obstacles’ or ‘blocks’ in the English subtitles). SOLA TS was insistent but now uncertain, asking again for HELGE INGSTADT to turn to starboard ‘if it is you’. By the end of this conversation, the ships are now 1 mile apart, and HELGE INGSTADT is still cruising at 17kts. They are 2m30s from collision.

There is then a sequence of poorly-translated messages where HELGE INGSTADT appears to suggest that she has ‘a few degrees starboard when we passed’. This could mean one of two things – either she has the flexibility to alter her course only a few degrees to starboard, or else she believes that she may be passing the SOLA TS on her starboard side and that a collision is not anticipated, although it would clearly be incredibly close. Whatever the intent, the effect is the same; a very close quarters situation is inevitable.

SOLA TS appears to be resigned to a collision and, after insisting that HELGE INGSTADT has to ‘do something’, states that a collision is essentially inevitable. Subsequent to this, less than 5 minutes after the initial radio hail and with HELGE INGSTADT still at speed, they collide. From the images of the HELGE INGSTADT, the impact of the tanker’s starboard bow on her hangar (midway down her starboard side) is clear, the anchor hawse gouging a deep hole along her hangar side. Crucially, however, SOLA TS was fully laden and her submerged bow has clearly split open the HELGE INGSTADT below the waterline.

The engine room is probably flooded, leading to the reported loss of power. HELGE INGSTADT rapidly loses speed and the actions of her bridge team are immediate, a hard turn towards the shore to run her aground. The speed of that decision is stunning; an immediate recognition of the gravity of the situation, and the only option to keep her afloat. Within 2 minutes of the collision, she is aground. It is possible she drifted ashore under her momentum alone; it is doubtful she had any motive power at this point. After the collision, HELGE INGSTADT requests ‘immediate assistance’ (and does not declare Mayday at any stage), reporting that she had hit an ‘unknown object’.

Note from the screenshot above just how close SOLA TS is to the other merchant vessels. SILVER FIRDA is just clear ahead, but SOLA TS makes a very substantial turn to starboard after the impact, resulting in a very close quarters situation with VESTBRIS who makes a bold alteration of course to port to bail out.  Further astern, both SEIGRUNN and the unknown radar trace open out to starboard, although the unnamed contact slows down rapidly, apparently catching SEIGRUNN unawares and the two have a very close quarters situation.

Additional Factors

Automated Identification Systems (AIS)

AIS is a civilian identification system adopted by the International Maritime Organisation (IMO) for the purposes of safety at sea, enshrined in the Safety of Life At Sea (SOLAS) convention. It is the maritime equivalent of IFF for civilian aircraft; mandated for all ships grossing more than 300 tonnes and for all passenger ships, it is a useful aid to navigational safety. It is only as good as the information transmitted, however – position, course and speed should be automatically relayed from the ship’s GPS system, but failures can lead to wrong information being transmitted, and supplementary information such as whether the ship is underway or moored requires manual intervention to change settings.

AIS is not a replacement for radar or a good lookout. The IRPCS mandate that a lookout must be kept ‘by sight and hearing, as well as all available means appropriate in the prevailing circumstances and conditions’ (Rule 5). This is reinforced by Rule 7, which not only restates the requirement to use ‘all available means appropriate’ to determine if risk of collision exists, but also drills into the use of radar information as well. Nowhere in the Rules is AIS mentioned, although it is clearly implies by the phrase ‘all available means appropriate’. The ‘so what’ is that navigation without AIS is inadvisable, but not incorrect. AIS data is usually plotted on an electronic chart and/or radar display, thus providing a combined picture of all available navigation information to the mariner.

Although most warships carry AIS systems, national caveats on their use are perfectly legitimate – after all, there’s no point in operating stealthily at sea if you broadcast your position. During TRIDENT JUNCTURE it makes sense for warships to operate with AIS switched off, especially if repositioning to set up particular attack phases or if trying to evade surveillance by the opposition. Such activity is routine in Royal Navy exercises including FOST training on a weekly basis south of Plymouth and during the twice-annual JOINT WARRIOR exercise off Scotland.

Nevertheless, it is difficult to understand why, in home waters and in this busy area, a transiting warship would elect not to switch on its AIS system.

Exercise TRIDENT JUNCTURE

TRIDENT JUNCTURE ran from 25 October to 23 November 2018. Taking place in three phases, the second phase was the ‘live’ phase which ran from 25 Oct to 7 Nov, therefore the HELGE INGSTADT collision took place the day after the live phase ended. The third phase was a command post exercise, not requiring live assets (more of a simulation than a live exercise). At this stage it is not known whether HELGE INGSTADT was still participating in the exercise; if she was, then running AIS silent through coastal waters is a valid tactical manoeuvre, and could potentially have been combined with a radar and emission silent policy; in open waters, a warship could even operate without lights, or with lights dimmed or rigged in ways to suggest she is in fact much larger or smaller than might otherwise be estimated.

However, this sort of tactical deception in narrow, confined and/or busy waters requires planning and perfect execution in order to be safe and successful. At minimum in an equivalent RN platform, it would have required a qualified navigator and a command-qualified officer (CO or competent XO) on the Bridge augmenting the standing watch team, with additional personnel closed up on the engines, steering and radar plots.

GPS Jamming

There has been much speculation as to whether GPS jamming, whether Russian or otherwise, could have force this incident.  The answer in my view is no on two counts.

First, the AIS data and radar plots remain correlated throughout.  If there was GPS jamming ongoing, we would expect to see the two data sources diverge.  In the absence of that, I do not believe that jamming was taking place.

Secondly, even if GPS jamming was ongoing, the fundamentals of navigation require that we use visual and radar information above all others; this forms the basis of bridge watchkeeper training.  This is unaffected by GPS jamming, and the situation would be clear to any and all vessels in the area.  The radio conversation clearly refers to an unknown vessel  initially, which was plotted on radar and was probably sighted visually as well; it could not have come from a GPS-derived AIS source.

Analysis

When considering collisions such as these, the first reference is always the IRPCS, the ‘Rules of the Road’. This situation under the IRPCS is a Head On Situation (Rule 14), that is ‘when two power driven vessels are meeting on reciprocal or nearly reciprocal courses so as to involve risk of collision’. Neither vessel benefited from any special status under any other Rule, and both vessels are required to ‘alter her course to starboard so that each shall pass on the port side of the other’.

SOLA TS is a large ship that is relatively unwieldy at low speeds. At these close ranges, her speed of reaction would have been substantially slower than that of HELGE INGSTADT; nevertheless, she was obliged to take action to avoid collision under Rule 14. However, SOLA TS was hemmed in to starboard by the overtaking group of merchant vessels; altering course to starboard was possible but extremely limited given the proximity of SILVER FIRDA and VESTBRIS. By itself, that manoeuvre would not likely have been sufficient to avoid the close quarters situation but it might have opened up enough of a gap for HELGE INGSTADT.

Turning the other way, however, was fraught with risk.  If HELGE INGSTADT obeyed her imperative under Rule 14 late and turned to starboard, a port turn by SOLA TS would have kept the ships on a collision course. Therefore SOLA TS was limited in her options, which probably explains her insistence that HELGE INGSTADT be the one to move. Although understandable, it does mean that SOLA TS is not absolved of all responsibility, as the ‘catch all’ Rule 2 of the IRPCS elegantly describes.  And if these constraints were weighing on the master of SOLA TS’ mind at the time, you might expect more forceful and detailed instructions over the VHF channel.

HELGE INGSTADT should have had more options available to her.  She was not confined by overtaking traffic, and under Rule 8(e), ‘if necessary to avoid collision or allow more time to assess the situation, a vessel shall slacken her speed or take all way off by stopping or reversing her means of propulsion’.  As a warship of her type, HELGE INGSTADT is expected to be more than capable of reducing her speed rapidly; indeed, she could have reversed course in the time available.  Since she declared she was unable to alter her course to starboard, a reduction in speed should have been the first option to be taken in order to allow more time, and later to avoid collision.  An alteration of course to port would have taken her across the bow of the northbound merchantmen, itself a risky manoeuvre but viable given the greater area of navigable water to the east.

Assumptions

• All ships were maintaining an appropriate lookout, visually and by radar.
• None of the ships involved were classified under any special condition under the IRPCS.
• All ships were in sight of one another.
• As this was at night, all vessels had the requisite navigation lights switched on at the mandated brilliance.
• There is no other radio traffic of significance other than that heard in the videos.  The channel concerned is assumed to be VHF channel 16, the international safety channel.

Contributing Factors

• HELGE INGSTADT appears to not be transmitting on AIS, reducing other ships awareness of her position.
• Fedje VTS is unclear about the identity of HELGE INGSTADT at the beginning of the incident, suggesting they were not fully aware of her position and therefore did not warn the SOLA TS or other traffic of her presence earlier.
• SOLA TS was at low speed, fully laden and under tug escort, but was otherwise not restricted in her ability to manoeuvre.
• HELGE INGSTADT was cruising at 17kts throughout and did not slow down prior to collision, even when obliged to do so under Rule 8 of the IRPCS.
• The merchant vessels SILVER FIRDA and VESTBRIS, transiting north, did not manoeuvre at any stage to allow SOLA TS room to alter to starboard as the close-quarters situation with HELGE INGSTADT develops.
• Communication between HELGE INGSTADT and SOLA TS does not make clear any limitations placed on SOLA TS in terms of speed, manoeuvrability or the proximity of overtaking traffic.

What we do not know.

What we do not know is as important as what we do.  We have to assume some of the information covered above is suspect or has been edited; nonetheless, much of it hangs together across three different sources.

We cannot be certain of the weather conditions which may have restricted visibility.  However, there is no evidence of weather or high seas on the radar picture and if visibility was restricted by rain or fog, a ship would be unlikely to be sailing at high speed.

We do not know if HELGE INGSTADT was operating under any particular defect or operational condition (such as having a towed sonar system deployed) that affected either her manoeuvrability or her speed.  However, none was declared over VHF.  We similarly do not know whether SOLA TS was operating under any special condition authorised under the IRPCS.

Conclusion

I cannot and will not assign blame to anyone in this situation, as I do not possess all the facts and I am using potentially questionable information.  However, I would have questions as to why HELGE INGSTADT, faced with a narrowing channel and 4-5 merchant vessels all on reciprocal or near-reciprocal courses, did not reduce her speed to allow time to either assess the situation fully and take action, or avoid collision.  The formal report will determine the correct facts of this case.

However, for all mariners I hope this is a useful study that could be useful when navigating in similar situations in future.  It is certainly something I will have my future watchkeepers study when they are keeping the ship safe overnight.