Night Deck Landing Accident While Training off Brazil (Omni Leonardo AW139 PR-OTF)

On 2 June 2021 Omni Helicopters Leonardo AW139 PR-OTF, under charter to Petrobras, was involved in an accident landing at night on the SS-75 Ocean Courage Mobile Drilling Unit (MODU) off Brazil.

Wreckage of Omni Leonardo AW139 on the Deck of Ocean Courage (Credit: CENIPA)

The copilot and an occupational-safety technician who was on the helideck suffered serious injuries.  The aircraft commander suffered minor injuries and one other occupant, also a pilot, escaped uninjured.

Diamond Drilling SS-75 Ocean Courage (Credit: Brazilian Navy)

The Brazilian safety investigation agency CENIPA issued their safety investigation report on 17 September 2024.

The Accident Flight

It appears that at 18:53 Local Time the helicopter departed Jacarepaguá – Roberto Marinho Aerodrome, Rio de Janeiro for the Ocean Courage with three occupants.  The sun had set at 17:21.  Departure was c 30 mins later than planned because the aircraft commander had become stuck in traffic en route to the airport.

Night offshore crew change flights are not permitted in Brazil.  However, night emergency flights for medevacs are allowed.  CENIPA indicate there was an exemption for night deck training flights and the operator did have approved night training procedures (albeit referencing a regulation repealed in 2018).

In the previous 2 years the operator had only done 15 night offshore landings, in contrast to 12,000 daytime offshore landings.  Due to the rarity of night flights only a small cadre of pilots were rostered for night standby.  CENIPA say that…

…there was an understanding within the company that carrying out a nighttime landing in the simulator would be sufficient to provide the pilot with the necessary qualification to conduct the real operation.

[At] the time of the occurrence, the ANAC did not determine a specific requirement for night-time operation in helicopters…

A further challenge to night training & recency was that Petrobras had not made offshore helidecks available for the 2 previous years.

This training flight had been postponed several days so the crew was not the one originally intended, which had included an instructor with greater night recency.  Both pilots were captains.

The aircraft commander, who was Pilot Flying, had 9,321 hours, 4,073 on type and was also an instructor. He had worked for the company for 11 years.  On 5 May 2021 he had completed a simulator session that include a night onshore landing.

There were comments in the pertinent records on the concept of stabilized approach, with special attention to night-time flights, including the possibility of spatial disorientation and optical illusion phenomena.  However, the flight records did not inform precisely what had occurred during the training.

The simulator instructor commented that the aircraft had operated the aircraft safely, complying with the profiles prescribed. His performance was considered “satisfactory”.  It does not appear that he had previously conducted an actual night offshore deck landing though.

The copilot, the Pilot Monitoring, had 4,863 hours, 3,395 on type. He had completed an onshore night landing in the simulator in January 2021.  There were suggestions he had completed 3 night deck landings in 2018 but according to CENIPA no evidence was found that confirmed that.

The third occupant was described as onboard to “change the pilot’s seats” when on the deck according to the CEIPA report. This was not elaborated on but is highly likely to be ‘lost in translation’ and they were in fact going to swap with the co-pilot to do their own night landing practice session next.

The only pre-flight briefing was conducted while walking to the aircraft.  It appears the crew did not have checklists onboard and that no approach brief was given.

The wind strength at the helideck was 13 kt, at a direction of 043° with no precipitation.

In relation to the bow of the platform, the wind direction was 263°.

The operator’s SOP 8 specified the ‘Traffic Pattern for Oil Rigs and Vessels’.  After identifying the installation, PR-OTF joined the traffic pattern described in the SOP…

…by the right and entered the final leg from the stern of the platform, continuing on the approximate heading of 056°, that is, heading up wind, but with a left component, since the wind direction, according to the METAR of the area, was 040°.

The commander chose this approach axis to allow landing on his side.

The final approach profile executed from the height of 300 ft, is shown five graphics (‘Points 1-5’ using a generic fixed installation and an EH101 to represent the MODU and AW139).  CENIPA consider the approach was normal at Points 1 & 2 (at 300 ft and 200 ft respectively).

At Point 3, the helicopter descended to 80 ft (below the helideck height of 98 ft).

In this condition, the aircraft maintained a magnetic heading of 060°, an indicated airspeed of 37 kt., and a rate of descent of 25 ft/min. The aural alert “150 ft” was also emitted, followed by “Check Height”.

This indicated an unstable approach.

It is noteworthy that no proactive action was observed on the part of the PM to alert the PF of the need to perform a go-around, given the aircraft profile outside the parameters required for a stabilized approach.

That being said, the phenomenon known as “Black Hole” may have occurred in this segment of the flight over the sea, since the platform lights were the only available light sources. Without peripheral visual references to help, orientation became more difficult, leading to the tendency to fly below the recommended profile for a normal approach.

Furthermore:

SOP 08 revision nº 06, dated 14 May 2021, in force at the time of the event, did not have a clear definition with well-defined criteria for a stabilized approach.

CENIPA continue:

At point 4… heading had changed to 096°, with a pitch angle of 29°. It was found that, at this point, the PR-OTF had a rate of climb of 432 ft/min.

The aircraft had now climbed from 80 ft to 205 ft.  The PM will have now lost sight of the installation and the PF’s view will have become more peripheral.

CENIPA indicate that at this point…

….SOP 08 advised that the pitch angle could not exceed 10°, and that heading variations should be limited to between 30° and 45° of the heading initially planned for landing.

During the landing attempt, the Vertical Speed (VS) reached -4,656 ft/min, the magnetic heading varied up to 145°, with a Pitch of 24°, yaw angle of 32°, and a roll of 10°.

CENIPA’s tabulated data indicates an altitude of 45 ft which would imply the aircraft then climbed to reach the deck.  Ground speed is shown as 24 knots vs target of 15 knots defined in SOP 08.

CENPIA depict the approach as follows (not necessarily to scale):

Then, in an already unstable condition and at a height above the one expected, the aircraft entered the helideck area and, in the transition to hovering flight, there was loss of control of the helicopter after ample variations in height, heading, and power.

CENIPA confusingly comment that “despite the attempted go-around, the PR-OTF turned to the right” though its not clear where they believe a go-around was attempted.  This final turn, to give a tailwind, appears to be after the loss of control and perhaps indicates a deliberate attempt to crash land on the deck.  The…

…main rotor blades initially collided against the riser crane operator’s cabin and subsequently against the service structure located on the right-hand side.

MRB Impact of Omni Leonardo AW139 on Ocean Courage (Credit: CENIPA)

CENIPA classify this accident using ICAO common taxonomy as CTOL (Collision with obstacle during takeoff and landing) but in practice this seems to have been a LOC-I (loss of control inflight) that then resulted in an obstacle collision.

The helicopter had flown just 67 hours 40 minutes since new.  The wreckage was destroyed in 2023 in a hangar fire.

Safety Investigation & safety Analysis

No evidence was found of failures or malfunctions in the aircraft systems and their components that might have contributed to the occurrence.

Similarly, there were no signs of any contribution from the SS-75 [Ocean Courage] infrastructure to the accident.

However, the Ocean Courage was not fitted with the latest ‘Circle and H lights‘, introduced after extensive trials as a regulatory requirement in the UK to give enhanced visual clues.

CENIPA hypothesise that the commander’s late arrival at the airport may have resulted in skipping a formal pre-flight brief and the failure to identify and source the missing checklists.

CENIPA identify the following contributory factors:

  • Attention: On the final trajectory towards the installation the focusing of the pilot’s attention on the approach course compromised his personal alert system, reducing his ability to respond effectively and accurately to the pieces of evidence that the helicopter was experiencing an unstabilized approach.  With regard to the flight dynamic, the absence of both the checklist and the QRH may have led to a decrease of the pilots’ attention, which led to lapses associated with automatic processes, since they failed to perform the Approach Briefing, a standardized procedure
    that should have been undertaken before the descent (Before Descent Checks).
  • Attitude: On the final approach for landing, even with the “150 ft.” and “Check Height” alerts being issued, indicating that the aircraft was below the height of the platform (80 ft.), no corrective action was taken toward initiating a missed approach. The crew decided to take off even though they had not found either the QRH or the checklist on board the helicopter.
  • Training: it was found that the operator [Omni] had difficulty scheduling night-time flights with the contracting company [Petrobras] due to the unavailability of platforms accommodating this type of mission.  In the years prior to the accident, the contracting company had suspended or postponed this type of training, which resulted in a period of approximately 2 years without this type of operation, revealing the inefficiency of the systematized processes aimed at improving the operational skills of the pilots. In addition, the training carried out in the flight simulator proved to be insufficient to provide adequate performance and efficiency in the context of this type of operation. [Note: there is no evidence CENIPA specifically evaluated the simulator training so this conclusion is unproven].
  • Crew Resource Management: During the whole landing circuit, no assertive action was observed on the part of the PM to alert the PF on the need to perform a go-around in the air, given the evidence observed in the context of an unstabilized approach.
  • Handling of aircraft flight controls:  The aircraft entered the helideck area in an unstable condition and at a height higher than expected and, in the transition to hovering flight, loss of control of the helicopter occurred, after wide variations in height, heading, and power.
  • Piloting judgment: The selected approach axis compromised the maintenance of visual references in the transition from the LDP to the landing spot on the helideck.  Likewise, there was a delay in the decision to go around, since the decision was made after the LDP.
  • Management planning: There was inadequacy on the part of the operator’s planning in relation to the allocation of [pilots to the] activity in question, given that an inexperienced crew was assigned to this type of mission without the presence of an instructor with prior training in real night-time landing on a maritime platform.
  • Insufficient pilot experience:  The pilots did not have the necessary experience to operate a helicopter without the supervision of a qualified instructor, in the circumstances experienced during the night-time operation [to an offshore installation].
  • Decision-making process: The crew had difficulty perceiving, analyzing, and acting appropriately in relation to the risks involved in landing at night on the platform, even after having experienced an unstable condition during the final trajectory. These difficulties led them to make an inadequate
    judgment about continuing the approach for landing on the maritime unit.

‘Undetermined’ factors listed by CENIPA include: disorientation, emotional state, illusions, motivation, perception, flight planning and ‘support systems’ (including procedures & regulations).

Safety Recommendations

Three safety recommendations are raised to the aviation regulator ANAC and one to the Department of Airspace Control (DECEA).

Safety Actions and the HeliOffshore Industry Action Plan for Night Deck Landing Practice

CENIPA list three safety actions:

After the accident, the operator conducted internal studies, and issued the SOP 43 – “Offshore Night-time Flight”, with the purpose of providing pilots with more details on nighttime operations and establishing shorter intervals for training flights.

An agreement was established between the offshore [helicopter] operators that have contracts with the contracting company [Petrobras] , so that the training process for night-time flights is made regular and effective.

On 23 March 2022, the ANAC published the Supplementary Instruction n° 135-005, Revision A, establishing procedures for Aeromedical Operations conducted by Air Operators under the RBAC-135.

It should be noted that 135-005 is a general air ambulance notice and does not address night offshore operations specifically.

More importantly, HeliOffshore have developed an “Industry Action Plan for Night Deck Landing Practice“, published in December 2022.  This action plan specifically followed this accident (‘Event 1’) and a serious incident (‘Event 2’) in another region where an excessive rate of descent developed and the helicopter narrowly missed impacting the water during night deck landing practice.  They identified the following causal factors:

HeliOffshore…

…raised 21 potential improvement opportunities, each of which require multi-stakeholder collaboration to deliver consistent, far-reaching and robust operational safety performance improvement at an industry level.

Our Observations

This accident illustrates the additional risk of night operations, particularly when only conducted for training and emergencies.

It is another accident that indicates the benefit of integrating Threat & Error Management (TEM) into briefings and operations.

As noted by HeliOffshore, Flight Data Monitoring (FDM) should be a means of feedback to reduce the risk of this sort of accident.  To do that not only so the right event parameters need to be defined but night flights need to me examined distinct from day flights too.  It is surprising that CENIPA don’t discuss FDM, something whose value was first demonstrated offshore nearly 25 years ago.

We are now viewing this accident in hindsight, a luxury that crews don’t have during a flight.  While CENIPA discuss human factors and CRM extensively, it is paradoxical that the ‘voice of the crew’ is absent from their report, so the reasons that the developing events still made sense at the time to this crew are unarticulated.  Without that insight the CENIPA analysis is primarily counterfactual.

Safety Resources

The European Safety Promotion Network Rotorcraft (ESPN-R) has a helicopter safety discussion group on LinkedIn.  You may also find these Aerossurance articles of interest:


Aerossurance has extensive air safety, flight operations, SAR, airworthiness, human factors, helidecks, aviation regulation and safety analysis experience.  For practical aviation advice you can trust, contact us at: enquiries@aerossurance.com