A333, Kathmandu Nepal, 2015

A333, Kathmandu Nepal, 2015

Summary

On 4 March 2015, the crew of a Turkish Airlines A333 continued an automatic non precision RNAV approach below the prescribed minimum descent altitude without having obtained any element of visual reference and when this was acquired a few seconds before the attempted landing, the aircraft was not aligned with the runway centreline and during a 2.7g low-pitch landing, the left main gear touched down on the grass. The aircraft then left the runway completely before stopping with a collapsed nose gear and sufficient damage to be assessed a hull loss. None of 235 occupants sustained serious injury.

Event Details
When
04/03/2015
Event Type
HF, RE, WX
Day/Night
Day
Flight Conditions
Not Recorded
Flight Details
Operator
Type of Flight
Public Transport (Passenger)
Take-off Commenced
Yes
Flight Airborne
Yes
Flight Completed
Yes
Phase of Flight
Landing
Location
Location - Airport
Airport
General
Tag(s)
Non Precision Approach, Deficient Crew Knowledge-automation, Deficient Crew Knowledge-handling
HF
Tag(s)
Inappropriate crew response - skills deficiency, Inappropriate crew response (automatics), Plan Continuation Bias, Procedural non compliance, Ineffective Monitoring - PIC as PF, Ineffective Monitoring - SIC as PF
RE
Tag(s)
Directional Control, Off side of Runway, Continued Landing Roll
WX
Tag(s)
Fog
EPR
Tag(s)
Aircraft Exit Injuries
Outcome
Damage or injury
Yes
Aircraft damage
Hull loss
Non-aircraft damage
Yes
Non-occupant Casualties
No
Occupant Injuries
Few occupants
Off Airport Landing
No
Ditching
No
Causal Factor Group(s)
Group(s)
Aircraft Operation
Safety Recommendation(s)
Group(s)
Aircraft Operation
Aircraft Airworthiness
Air Traffic Management
Investigation Type
Type
Independent

Description

On 4 March 2015, an Airbus A330-300 (TC-JOC) being operated by Turkish Airlines on a scheduled passenger flight from Istanbul to Kathmandu touched down at a higher than normal rate of descent at destination in very poor day visibility with the landing gear partly on the grass before leaving the runway completely. The aircraft was extensively damaged and subsequently declared a hull loss but only one minor injury to the 235 occupants was recorded to a passenger during the evacuation. Impact damage was caused to lighting and signage and there was superficial damage to the surface of one of the two taxiways crossed by the aircraft during its excursion.

The extensively damaged accident aircraft where it came to a stop (reproduced from the Official Report)

Investigation

An Investigation was carried out by the Nepal Aircraft Accident Investigation Commission. Data from the FDR and 2 hour CVR were successfully downloaded.

The Investigation found that the 55 year-old Captain of the aircraft had 1456 hours flying experience on the accident aircraft type out of a total of 14942 hours, having joined the airline as a Co-Pilot eight years previously from another operator after acquiring his initial flying experience with the Turkish Air Force. The 47 year old Co-pilot had also gained his initial flying experience with the Turkish Air Force and had joined the airline 3½ years earlier after operating as an ATR 72 Co-Pilot. He had accumulated 7659 hours total flying experience which included 1269 hours on the A330. The Captain had never previously flown to Kathmandu but both pilots had received "Special Purpose Training" for Kathmandu in a simulator just over 2 months prior to the accident. The Co-Pilot had made two previous non precision approaches to Kathmandu but the accident approach was his first RNP approach. It was noted that authorisation for Turkish Airlines to fly RNP AR approaches into Kathmandu had only been granted a little over two months before the accident.

It was established that on initial contact with Kathmandu, two minutes after the airport opened at 0600L, the flight had been advised that the visibility was 100 metres and the airport was closed. After holding at FL210 until 0650L, an updated visibility of 1000 metres was notified and clearance was given for a RNP 0.3 AR Approach to runway 02. Three minutes after receiving landing clearance, a missed approach was commenced due to lack of visual reference. Fifteen minutes later, on receipt of updated weather information giving the visibility as 3000 metres, 1000 metres towards the south east and the lowest cloud as FEW at 1000 feet, a second RNP AR approach to runway 02 was commenced. Two minutes after giving clearance to land and with the aircraft passing 880 feet aal at the time, TWR asked if the runway was in sight and were advised that it was not. Passing the prescribed MDA, the crew asked if the approach lights were on and were advised that they were at full intensity. The aircraft was found to have then continued the approach with the AP engaged (expressly not permitted for the final, visual, stage of an RNP AR approach) and it remained engaged until 14 feet agl, at which point it was disconnected and a flare attempted. The aircraft touched down with the left main gear assembly on the grass 10 metres to the left edge of the 3000 metre long 45 metre wide runway paved surface with a recorded 2.7 g vertical acceleration. It initially continued in a straight line before the soft ground beneath the left main gear caused the aircraft to leave the runway altogether to the left. It came to a stop wholly on the grass with a collapsed nose gear and with substantial damage to both the fuselage and the engines.

The evacuation in progress (reproduced from the Official Report)

The crew requested fire and medical assistance in response to a query from the TWR as to whether they had landed. AFS vehicles located the aircraft and three of them carried precautionary applications of water and foam in the presence of a fuel leak and 'smoke' coming from both engines. After a delay, an emergency evacuation was completed.

Circumstantial evidence indicated that the visibility had quickly and materially deteriorated from about the time landing clearance had been given. It was described by a domestic flight crew taxiing for departure at the time of the crash as "almost zero", a description supported by CCTV recordings and reports from military personnel located near to the threshold of runway 02. The Investigation concluded that the visibility at the time of the accident had been 200 metres in fog - much worse than at the time of the earlier go around. It was noted that the deteriorating trend had not been communicated to the crew by TWR and neither had a SPECI for this change been issued by the Meteorological Office.

The explanation for the off runway centreline position of the aircraft at touchdown - apart from continuing through MDA and almost to the runway without visual reference - was found to be the incorporation of an AIP Supplement which incorrectly showed the runway 02 threshold co-ordinates offset about 26 feet to the left into the FMS of the accident aircraft; although the actual runway threshold had never changed. This information had been generated by CAA Nepal personnel as part of a notification of a displaced threshold which it was planned to introduce to facilitate runway extension work. A NOTAM issued on the effective date of the one-month-validity AIP Supplement to defer the introduction of the displaced threshold was not actioned by the chart provider LIDO and the displaced threshold (and erroneous offset) version of the approach chart used therefore remained in the FMS. It was noted that, although the Istanbul flight had been operating into Kathmandu with the displaced and erroneously offset threshold location loaded into the FMS for almost a month, the discrepancy had only been first reported by a crew who had operated into Kathmandu two days previously. And because action on this report was still pending on the day of the accident, the accident aircraft crew, whilst briefed pre-flight on the NOTAM cancelling the displaced threshold, would not have known that their LIDO Chart was (still) based on the displaced threshold and erroneously-offset threshold position. Nevertheless, it was noted that for any crew visual at or before the prescribed MDA, the minor realignment of the flight path necessary at that point would have been readily accomplished.

It was found from the CVR recording that:

  • After the missed approach a member of the cabin crew had stated to the Captain their concern at the prospect of diverting to Delhi
  • Immediately after this, the First Officer had told the Captain that the RNP approach would "bring them directly to the runway"
  • Four seconds before the automatic annunciation of 'MINIMUM' the First Officer had said "it will appear if we descend below"
  • In response the 'MINIMUM' call, the Captain said "continue until 300 feet" contrary to the SOP response from the PF (if visual) of "visual and continue"
  • There was no call of "visual" from PM who was required under SOP to look out for visual reference whilst the PF stayed on the instruments.
  • Approximately 5 seconds before touchdown, the PF stated "appearing" - the first reference to the runway being in sight.

It was concluded that "the aircraft continued its approach below MDA without the proper visual reference contrary to the standard and procedure of the RNAV(RNP) approach" and noted that it had also done so contrary to the requirement that on such an approach, the AP must not be used below MDA.

The Investigation formally determined that "the Probable Cause of this accident is the decision of the flight crew to continue approach and landing below the minima with inadequate visual reference and not to perform a missed approach in accordance to the published approach procedure".

It was also determined that there were two Contributory Factors as follows:

  • the probable fixation of the flight crew to land at Kathmandu
  • the deterioration of weather conditions that resulted in fog over the airport reducing the visibility below the required minima.

Safety Action taken as a result of the accident by Turkish Airlines and advised to the Commission included the following:

  • A unit which will quality control charts prepared by their service provider has been established
  • The minimum visibility required to conduct the Kathmandu RNP AR approach has been increased from 900 metres to 1800 metres
  • The rostered crew for flights to Kathmandu has been increased to three pilots
  • The scheduled landing time for the winter period had been changed

A total of 21 Safety Recommendations were made as a result of the Investigation as follows:

  • that Turkish Airlines should review the pilot qualification requirements to operate to and from Kathmandu.
  • that Turkish Airlines must ensure that the crew strictly adheres to the Standard State Instrument Arrival procedures and to the Airline's Standard Operating Procedures.
  • that Turkish Airlines must ensure that the correct navigation data have been uploaded into the FMGS NAV database of their aircraft.
  • that Turkish Airlines should have a system in place to act efficiently and effectively with full understanding of its gravity upon receiving the information of operational significance such as NOTAM and feedback of the crew etc.
  • that Turkish Airlines should establish a system of verifying the quality of charts prepared by the service provider.
  • that Turkish Airlines should establish a system of checking the validity of FMS database.
  • that Turkish Airlines should review its Kathmandu RNP AR Company visibility minima keeping in mind its own requirements over and above those of State-published visibility minima.
  • that Turkish Airlines should review its crew composition requirements to and from Kathmandu airport keeping in view of the flying time and time zone etc.
  • that Turkish Airlines should ensure that the crew strictly follows the safety related procedures and cockpit discipline.
  • that CAA Nepal should review its requirement published in the State AIP regarding crew qualification before they are authorised to operate to and from Kathmandu airport.
  • that CAA Nepal must ensure that there exists an effective and efficient coordination between aeronautical information services and aerodrome authorities.
  • that CAA Nepal must ensure that raw aeronautical information/data are provided by the aerodrome authorities take account of its accuracy and integrity requirements for aeronautical data as specified by ICAO Annex 15 and its Aeronautical Information Service Manual.
  • that CAA Nepal must ensure that there exists a proper planning for works to be accomplished before disseminating such information through Aeronautical Information Services with full understanding of its gravity.
  • that the Kathmandu Airport Met Office must ensure that it disseminates SPECI reports when visibility deteriorates as prescribed in ICAO Annex 3.
  • that the Kathmandu Airport Met Office should have a system capable of providing a weather observation immediately after an accident has occurred.
  • that CAA Nepal must ensure that Air Traffic Controllers on duty at Kathmandu Tower are vigilant and weather information representing deterioration in visibility minima are provided through them to the aircraft immediately.
  • that CAA Nepal should provide refresher training to all Air Traffic Controllers at regular intervals.
  • that CAA Nepal should include an ATIS status check in its Daily Facilities Status Check List Reporting Form at Kathmandu.
  • that CAA Nepal should restore the ATIS Communication facility at Kathmandu immediately.
  • that CAA Nepal should keep track of the AIRAC update cycle when cancelling an AIP supplement.
  • that Commercial Chart Provider LIDO should put in place a more robust system to check NOTAMs and act accordingly .

The Final Report of the Investigation was completed on 6 October 2015.

Related Articles

SKYbrary Partners:

Safety knowledge contributed by: