Learning the lessons – the loss of the Norwegian frigate Helge Ingstad

In 2021 the Accident Investigation Board – Norway (AIBN) published a detailed report covering the loss of HNoMS Helge Ingstad after she collided with an oil tanker in November 2018. There is much to be learned from this event that is applicable to the RN and global navies. In this in-depth, although far from exhaustive, article we describe the incident and look at some of the key lessons.

The accident in summary

On 8th November 2018, the frigate Helge Ingstad (HING) was heading south down Hjeltefjord returning to her home base at Haakonsvern, near Bergen after participation in NATO exercise Trident Juncture. Just after 0400, HING collided with the bow of the Sola TS, a fully laden 112,939 DWT tanker, heading north having just departed from Sture oil terminal. The bulbous anchor hawse pipe on the tanker’s starboard bow ripped a large gash down the aft starboard side of the frigate. As HING heeled slightly under the impact, the tear in the side was extended sufficiently below the waterline to cause significant flooding.

Fortunately, there were no fatalities but HING suffered a brief total electrical failure after the impact. Power was partially restored but confusion and internal communication problems limited damage the control effort. Having lost steering and engine control, the ship ran aground on rocks about 10 minutes after the collision. Believing the flooding made capsize a likely, the crew abandoned ship in an orderly manner, assisted by tugs and rescue craft. Despite efforts to pin the ship on the rocks by tugs and with steel hawsers, HING progressively flooded sinking until almost entirely submerged.

Some months later, after complex clearance diving operations to remove fuel and live weapons and then secure chains beneath the wreck, HNIG was raised off the sea bed using a heavy lift crane barge. The wreck was eventually placed on a semi-submersible transportation barge and taken to Haakonsvern for further investigations and to remove sensitive material. Temporary patches were applied to the damaged shell plating and the ship was refloated. Unsurprisingly the navy declared the ship a total loss and in Jan 2021 handed her over to be scrapped, a process that is ongoing at the time of writing.

The cost of the accident is estimated at $80m for the salvage, operation, not to mention enormous reputational damage. Building a direct replacement for a ship that ceased production a decade ago is estimated at $1.4 billion and not considered viable. The Norwegian Navy was already overstretched, primarily occupied with monitoring its aggressive neighbour, lost 20% of its frigate force.

Around the world and in Norway particularly, there was considerable shock that this could happen, especially involving a warship operating close to home and in waters monitored by a traffic management system. Some hasty judgements were made by those speculating about the event, with some suggesting the accident “must have been the fault of the tanker crew as a highly trained naval personnel on board a modern warship with sophisticated sensors would be unlikely to make such a mistake.” Some also assumed the subsequent sinking was because: “the ship must have been poorly constructed by the Spanish shipbuilder Navantia as warships are specifically designed to survive damage and flooded compartments”. The investigation has proved neither of these assumptions to be the case.

Standing into danger

The root causes of the accident were entirely human error, primarily on the bridge of the warship. It was a clear and calm night and there were no technical problems that contributed to the initial collision.

HING was proceeding down the Fjord at a brisk 17–18 knots and was not transmitting her name and position on AIS. The Fedje Vessel Traffic Service (VTS) which managed shipping movements in the area was notified of the frigate’s arrival an hour and 23 minutes before the collision but the operator failed to plot her on the tracking system. Sola TS told the VTS when she left the oil terminal but the VTS operator had forgotten about HING and initially did nothing. The tanker was showing standard navigation lights but also had powerful floodlights illuminated to assist crew at work on the deck as was their standard practice.

The Norwegian Navy (Sjøforsvaret) was short of qualified navigation officers and the officer of the watch (OOW) in charge of navigating the ship was young and relatively inexperienced. The frigate had a crew of 137 on board, the majority were asleep at the time of the incident. 7 personnel were standing on watch on the bridge, including the OOW, an officer and a rating under training and 4 other ratings.

The OOW and trainee officer discussed the floodlights but believed they were ashore and stationary. Confirmation bias and lack of experience led the OOW to keep thinking this was the case until too late. Inexplicably there was only brief use of navigational radar and AIS to check the situation and they interpreted the tanker as a stationary object alongside at the terminal. There was a reliance on visual cues for too long and ironically the accident would probably have never happened if the visibility had been poor.

The pilot on the tanker saw HING in the distance and called VTS asking if they knew what this unknown vessel was. VTS was initially unable to help, having forgotten about HING. The tanker tried to signal the frigate by flashing lamp but it was not seen, probably obscured by the deck lights. The VTS operator then remembered HING and immediately informed the pilot on the tanker of her presence. The pilot called the frigate by VHF radio requesting an immediate turn to starboard. The Pilot had not clearly identified which ship was calling and the OOW thought he was speaking to another vessel approaching on their port side. Despite now being much closer, he still thought the lights were stationary and on the shore and there was no space to make a turn to starboard. VTS, which held both vessels clearly on radar, did not intervene to tell HING to stop, assuming the two ships, now in contact, would resolve the issue.

With the ships just 500m distance apart, the tanker stopped engines and radioed HING to take urgent avoiding action. At this point, the OOW finally realised the lights were a moving vessel and ordered a turn to port followed by rudder amidships, attempting to steer around the tanker. Unfortunately, the turn was made just a few seconds too late and a collision was unavoidable. Contact between the ship lasted 5 seconds and tore a 46m gash down the side, severing cables, pipes, control panels, switchboards and watertight bulkheads and doors.

Losing the ship

In the moments after the impact there was considerable confusion on the bridge. The team was trying at work out what exactly had happened and did not know how seriously the ship had been damaged or whether the crew had suffered fatalities and injuries. Stress and fear were factors in a situation that was beyond what they had trained for. There was a 10 second total power failure and it took several minutes for many systems to be manually restarted.

The Integrated Platform Management System (IMPS) which is the key technology used to control and monitor lean-manned modern vessels showed 564 alarms but this mass of information did not indicate the priorities for action. At the same time, they lost steering and engine control on the bridge and were unable to communicate with the Machinery Control Room (MCR) which also acted as HQ1 damage control centre.

Down below the sleeping crew were suddenly jolted awake but most had no idea what had happened. The junior rates mess on 3 deck was flooded and some officers were trapped in cabins in 2 deck, fortunately without serious injury but needing assistance to escape. The CO was asleep in his cabin behind the bridge and was thrown from his bunk by the impact. Dazed, he went quickly to the ops room and then to the bridge. The frigate was now heading towards the shore are about 5 knots with no control possible from the bridge. The steering gear was in full working order and the sailor on watch was contacted by sound-powered telephone but did not understand what was said. Realising they would soon run aground, the OOW called “full astern”, over the main broadcast. This was not heard by the propulsion controller or those in the aft main engine room. The OOW ordered the anchor be dropped but it was too late and finally “brace, brace, brace” was called before the ship hit the rocky shore almost directly bows-on at 0411, smashing the bow-mounted sonar dome.

Watch changes had just taken place in the MCR/HQ1 just before the accident and some of the relieved team were still around. They knew something serious had happened, possibly an engine or electrical switchboard explosion but observing damage and flooding, then thought they had run aground. Personnel were sent to inspect the lower compartments and eventually confirmed flooding in three separate watertight compartments – the after generator room, junior rates mess and storerooms. Some attempts at shoring and Damage Control were made in the generator room but the hole was large, obscured by pipes and live electrical cables. The loss of saltwater main pressure reduced the effectiveness of the already inadequate bilge pump system which failed to remove any water from the ship. Many hatches and doors that would maintain watertight integrity were left open, essentially to facilitate movement of personnel and portable pumps which proved ineffective.

The electronic tool used to calculate the stability of the ship (which the Norwegian Navy had previously agreed was not fully fit for purpose) indicated that if a fourth compartment were lost the ship would become dangerously unstable. It was then reported that water was flooding into the reduction gear room through hollow propellor shafts and the command team now believed abandoning ship was their only option. Tugs has arrived and were attempting to pin the ship against the shore but it was possible she could slip rapidly into deeper water. The MEO concluded that potentially all 6 aft compartments would flood and the command team decided to abandon ship.

Crucially, before leaving it was discussed whether to close the watertight hatches and doors on 2 deck but the MEO considered it too risky to go down into the vessel again. The ship was evacuated by 0632 with the Captain the last to leave. Post-incident analysis showed that this was a mistake – there was time and the open hatches were the reason the ship eventually sank completely. The grounding and the leaking shafts were not the reason the ship sank as there still would have been sufficient buoyancy if the forward compartments to keep her afloat, had they been sealed.

The controversy over the leaking shafts can be explained in simplistic terms by a change made to the Nansen class from the original Spanish F100 parent design relating to a reduced underwater acoustic signature. The ships use controllable-pitch propellers, the pitch being set by hydraulic oil pumped down the shafts. To meet requirements for strength and shock loads, it was decided to install a hollow intermediate shaft between the oil pump box in the aft generator room through the aft main engine room to the flexible coupling in the gear room. The pump openings into the hollow shafts compromised watertight integrity but this was not noticed during the design and construction by Navantia and the subsequent classification process conducted by DNV in 2014. The Norwegian Defence Ministry did not cover itself in glory by suing the classification society DNV for $1.7Bn in damages but was forced to drop the case when the investigation showed the ship ultimately sank due to the failure to close internal doors.

Lessons

It would be wrong to entirely blame the inexperienced OOW and poor performance of the bridge team for the incident, the investigation showed there were management failures that extended to the highest levels of the Navy. The pressure caused by the operational tempo, something the RN and USN would recognise, resulted in inadequate training time and personnel lacking sufficient experience in relation to their responsibilities.

The AIBN reports issued 15 recommendations relating to ship navigation and 28 recommendations relating to damage control and ship systems. Key findings were improvement in training and competence for bridge teams, that warships always activate AIS when navigating in confined waters with other vessels (something the RN does not do consistently). Unlike RN vessels, HING was not fitted with a Voyage Data Recorder (VDR) used for post-incident analysis and this should be remedied. The majority of recommendations to the navy related to training, especially damage control in complex situations, competence in shutting down and sealing compartments and understanding, operating and maintaining safety-critical systems, especially bilge pumps.

HING passed her final FOST inspection at Devonport in March 2018 but 37% of the crew had changed in the intervening 8 months before the incident. The crew all said they had benefited greatly from the FOST programme and their own exercises but they had not practised for anything resembling these circumstances. The simultaneous failure of several technical systems, time pressure, significant flooding, loss of communication and the fact that this took place in the early morning proved overwhelming. Lean-manned warships reliant on commercially-derived IPMS and automation perhaps lack the number of watchkeepers needed for effective damage control. The use of technology for decision support and remote control of key systems implies these tools need to be extremely robust and programmed to provide fast and simple advice for a very wide range of scenarios.

The AIBN report concludes that “The collision resulted in severe damage to the vessel, over and above what she was designed to withstand”. The damage was significant but nothing like what potentially could be caused by a weapon impact. Despite the very heavy investment in warship survivability measures, this makes one wonder just how little damage modern combatants can sustain without being immediately crippled. The ship was damaged only on one side, yet suffered major internal communications failures. In is unclear why ‘runners’ were not sent more quickly more to pass messages to and from the bridge, HQ1, Ops room and damaged areas. HING was extremely lucky in many ways to suffer no fatalities. The collision accidents involving USS Fitzgerald and USS John S. McCain the previous year killed several sailors, although neither ship sank.

There is much more to learn from the report and the Norwegian Navy must take some credit for being reasonably open about events and the outcome of the investigation. Soon after the accident, lessons were shared directly by the Norwegians with the RN’s FOST organisation which has influenced its training methods and more widely will make a valuable contribution to the preparedness of NATO warships.