‘Procedural Drift’: Lynx CFIT in Afghanistan

‘Procedural drift’ in the operating detachment was noted by the Service Inquiry into a British military helicopter accident in Afghanistan that killed 5 service personnel.  This drift is said to have allowed standards to dilute during sustained operations in theatre.

The Military Aviation Authority (MAA) released the Service Inquiry (SI) report on 16 July 2015, into loss of the British Army Air Corp AgustaWestland Lynx AH9A helicopter ZF540 in a Controlled Flight Into Terrain (CFIT) accident in Afghanistan on 26 April 2014.

Wreckage of Lynx AH9A ZF540 (Credit: MAA)

Wreckage of Lynx AH9A ZF540 (Credit: MAA)

This was the first fatal loss of a British helicopter in 13 years of operations in Afghanistan (although a third party fatality had been caused by a Chinook in 2010).   It is also the first Lynx AH9A accident since the first flight of the re-engined variant in 2009 (we provide background on the AH9A below).

The Accident

The accident occurred during a two-ship vehicle interdiction training mission in the ‘Bowling Alley’ valley on the ‘Texas Helo Range’ range 20 km south of Kandahar Airfield, at a density altitude of 5400ft.  The helicopter made a rapid descent, assessed at between 2000ft/min and 2500ft/min,  from 400ft above ground level on its final pass.  It impacted the ground ‘wings-level’, ~5º nose up, with minimal yaw and “significant forward speed”, leaving a 75m debris trail.  The main fuel tanks were compromised and a post-crash fire ensued.

ZF540 Debris Trail (Credit: MAA)

ZF540 Debris Trail (Credit: MAA)

 

The Investigation

The SI notes that during the flight (including the last two passes), the military passengers, who were gaining experience of the AH9As operational role, were allowed to fire a General Purpose Machine Gun (GPMG) door gun.  The SI Panel comment that this had not been discussed in sufficient depth at the pre-flight briefing and that this part of the sortie appeared more ‘ad hoc’.  They also comment on that due to a shortage of Suitable Qualified and Experienced Personnel (SQEP) the commanders authorised each others flights rather than an independent flight supervisor conducting the authorisation.

The SI postulates, based on some limited Cockpit Voice Recorder (CVR) data, that a rapid descent may have been deliberate to “provide a thrill for the passengers”, though it is important to note they do also comment positively on the general airmanship and professionalism of the flying.   They also note a possible self induced time pressure.

ZF540 Final Pass (Credit: MAA)

ZF540 Final Pass (Credit: MAA)

The report notes that the entry point on the final pass was different, giving the crew different visual clues.   The SI Panel comments that the crew may also have been distracted trying to establish visual contact with the second Lynx.  Additionally they note that the radar altimeter (RADALT) was not being used effectively as the ‘bugged’ setting was too low to initiate a recovery in a descent (and was being routinely set lower that required by the standard operating procedures without challenge).

The report highlights the pilot got limited sleep (less than 6 hours) the night before and was possibly fatigued.

The SI Panel highlighted that issues identified by a standards inspection 10 months earlier (such as under-manning, individual ‘overstretch’ from repeated tours and the quality of training supervision and delivery) were still evident.

They also identified lower than expected occurrence reporting from the detachment compared to the rest of the Lynx AH9A fleet and other helicopters detached to Kandahar.  A previous Service Inquiry on a loss of control incident involving a Royal Air Force Airbus A330 Voyager made a similar comment about the lack of reporting of lost articles in the RAF transport fleet.

Procedural Drift / Practical Drift

In relation to the underlying environment of ‘procedural drift’ the SI Panel comment this occurs gradually over time.  Such drift occurs when group norms and practices start to deviate from formal procedures. In some cases this may be because procedures no longer match operational circumstances, however practicality can be a factor too.


One major study of an accident that featured drift was by Scott Snook, then of the US Army.  His book, Friendly Fire, examined the accident shoot down of two US Army Black Hawk helicopters on a peacekeeping mission in Iraq in 1994 by the US Air Force, and what he called ‘practical drift’.  In that case Snook highlights that drift can occur if procedures are initially overly complex. These are then relaxed to be practical but then gradually over-relaxed until an accident.  Snook warns however that sometimes a vicious circle starts after an accident when new complex rules are introduced, setting up the conditions for further eventual drift.  Drift can also creates inconsistencies in understanding (i.e. misinformation) across interfacing organisations.

There is a view, as for example expressed by Sidney Dekker, that in practice procedures must be interpreted by front-line personnel and that non-compliances should be viewed in context of the circumstances the crew find themselves.  In The Field Guide to Understanding Human Error Dekker lists several potential reasons for procedural drift:

  • Rules or procedures are over-designed and do not match up with the way work is really done.
  • There are conflicting priorities which make it confusing about which procedure is most important.
  • Past success (in deviating from the norm) is taken as a guarantee for safety. It becomes self-reinforcing.
  • Departures from the routine become routine. Violations become compliant behaviour with local norms.

UPDATE 13 September 2015: We examine another case of procedural drift after the US National Transportation Safety Board (NTSB) hold a Board Meeting on this accident: Gulfstream G-IV Take Off Accident & Human Factors

UPDATE 11 September 2016: Another case were a last minute change of plan was partly responsible for an accident: Final Report: AS365N3 9M-IGB Fatal Accident


Convening Officer’s Comments

The convening officer of UK military aircraft SIs is now Air Marshal Dick Garwood as Director General of the Defence Safety Authority (DSA).  The DSA, of which the MAA is now a part, was formed on 1 April 2015. Among his comments are the following:

I therefore support the Panel’s findings that the accident was a controlled flight into terrain event caused by the aircraft being established in a descent from which it was not fully recovered prior to impact with the ground…

Of note, the accident occurred during a period of reduced operational tempo as the campaign approached its culmination, and there was no undue operational pressure on the Lynx Detachment at the time…

The difficult question as to why a competent and experienced crew, on an excellent weather day (with the sun behind and only a light headwind), would inadvertently fly their serviceable aircraft into the ground is compounded by the fact that they did not recognise their impending situation until just before the aircraft impacted the ground.

Whilst no single factor led to this accident there were disappointing aspects, including planning, briefing, authorisation, supervision, currencies, training, and adherence to checks and procedures.

There is much to learn form this tragic accident…

Recommendations

The SI panel made a 33 recommendations including on:

  1. Use of RADALT
  2. Flight authorisation and supervision
  3. Reviewing of roles and responsibilities for senior commanders
  4. Reviewing of regulations for passenger carriage (RA2340 was revised after this accident)
  5. Reviewing the ad hoc seat used by aircrewmen
  6. Reviewing occurrence reporting
  7. Fitment of a solid state CVR (surprisingly the AH9A uses a magnetic tape recorder)
  8. Introduction by the MAA of a Regulatory Article to require retro-fit of Flight Data Recorders

UPDATE 20 August 2015: The Coroner’s Inquest is due in November 2015.  At a preparatory hearing the families had their request to hear the CVR refused by the MOD, who cited the need to retain such recordings for the exclusive use of accident investigators.

UPDATE 5 January 2017: The MAA have issued a Notice of Authorised Amendment (NAA 17/01) that introduces a revised Regulatory Article (RA) RA1205: Air System Safety Cases at Issue 5.  Among the changes is a new Annex A on Collision Warning Systems (CWS), flight recorders and Flight Data Monitoring to be addressed by Duty Holders in their ASSCs.  the change follows “recommendations of Lynx and Tornado Service Inquiries”.

Lynx AH9A Background

The Lynx AH9A (or AH Mk9A) utility helicopter was upgraded specifically for Afghanistan as part of an Urgent Operational Requirement (UOR).

M3M Fitted to a Lynx AH9A (Credit: Cpl Barry Lloyd RLC)

Twenty two AH9 helicopters were fitted with LHTEC CTS800-4N turboshaft engines (40% more powerful than the previous Rolls-Royce Gem 42), updated displays, door mounted guns (including a 0.5″ M3M to supplement the GPMG) and improved surveillance equipment. The type will be replaced by the AgustaWestland AW159 Wildcat AH1.


Aerossurance sponsored the 2017 European Society of Air Safety Investigators (ESASI) 8th Regional Seminar in Ljubljana, Slovenia on 19 and 20 April 2017.  ESASI is the European chapter of the International Society of Air Safety Investigators (ISASI).

Aerossurance is pleased to be supporting the annual Chartered Institute of Ergonomics & Human Factors’ (CIEHFHuman Factors in Aviation Safety Conference for the third year running.  We will be presenting for the second year running too.  This year the conference takes place 13 to 14 November 2017 at the Hilton London Gatwick Airport, UK with the theme: How do we improve human performance in today’s aviation business?ciehf 2017


Aerossurance has extensive civil and military air safety, operations, regulation and safety analysis experience.  For aviation advice you can trust, contact us at: enquiries@aerossurance.com