A343, Nairobi Kenya, 2008
A343, Nairobi Kenya, 2008
On 27 April 2008 an Airbus A340-300 crew lost previously-acquired visual reference in fog on a night auto ILS into Nairobi but continued to a touchdown which occurred with the aircraft heading towards the edge of the runway following an inappropriate rudder input. The left main gear departed the paved surface and a go around was initiated and a diversion made. The event was attributed to a delay in commencing the go around. No measured RVR from any source was passed by ATC although it was subsequently found to have been recorded as I excess of Cat 1 limits throughout.
On 27 April 2008 an Airbus A340-300 being operated by UK-based Virgin Atlantic AW on a scheduled passenger flight from London with the First Officer as PF had carried out a night auto ILS approach to Runway 06 with the A/T engaged and been cleared to “land at your own discretion” with visibility reported as 3000 metres and the W/V 050 at 5 kts. The crew stated that they became visual with the runway at a height of between 300 and 200 ft. and at the DH of 200 ft, both pilots had more than the minimum visual reference required and could see “all the approach lights and a good section of runway lights”. The autopilot was disconnected at 100 ft radio altitude and the PF began to flare the aircraft at between 75 and 50 ft radio altitude. The aircraft floated at around 20 ft for a few seconds before it entered an area of fog and the PF lost sight of the right side of the runway and the runway lights. The commander also lost sight of the right side of the runway.
The aircraft touched down in a normal attitude but on the main gear only; the body and nose gear did not contact the ground throughout the event. The PF was not aware that the aircraft was moving laterally on the runway, but the commander became aware of the left runway edge lights moving rapidly closer to him before he lost the lights completely and was only aware of their position by the glow of the lights illuminating the fog. The commander called “go-around” and the PF immediately advanced the thrust levers from idle to full thrust. The aircraft became airborne after less than five seconds on the ground, the gear retracted normally and the crew continued with the go-around, climbing to an altitude of 9,000 ft to enter the hold. During the ground roll the crew had heard and felt a rumbling and suspected that the aircraft might have departed the left side of the declared runway although they did not believe that the aircraft had left the paved surface. The aircraft entered the hold while the crew considered their options. Having decided to divert to Mombasa the commander informed ATC that they may have run off the runway side and that they wished to divert to Mombasa. The First Officer remained as PF for the diversion, which was followed by a normal, day Visual Meteorological Conditions (VMC) landing.
At an early stage of the investigation, in July 2008, the UK AAIB had issued a Special Bulletin to publicise factual information available at that time. This information was incorporated in the Final Report
The Investigation was carried out by the UK AAIB after the Ministry of Transport (Air Accident Investigation Department) of Kenya delegated the entire investigation to them and appointed an Accredited Representative to assist with the subsequent enquiries.
It was found that after loosing visual reference, during which time the PF had made a left rudder pedal input, ground markings showed that the aircraft had touched down heading away from the 45 metre wide runway centerline until the left main landing gear passed over a runway light (which had been destroyed) before continuing off the paved surface before curving right to run approximately parallel with the runway for 180 m. A set of parallel set of marks from the right main gear were found to not quite leave the paved surface (stopping 5 cm from the edge of the paved shoulder) although they were off the declared runway surface. Only very minor damage, within Aircraft Maintenance Manual limits, was found to the aircraft and consequential ground installations damage was limited to the one broken runway light.
It was noted in the Report that “the loss of visual references during the flare is a complex event. Sudden changes in RVR can occur due to the natural variability in the density of fog. The phenomenon of rapidly-forming drifting fog during the wet season at Nairobi is not fully understood. In addition, although advisories to pilots caution about such phenomena for certain times of the year it can occur outside those periods, depending on the climatic conditions. A modern instrumented RVR system capable of immediately displaying changing visibility was installed at NBO 18 months before this incident. However, its value was limited as, due to the absence of appropriate training on the AWOS, information from that system was not passed to flight crews and pilot-assessed visibility from several minutes earlier was routinely relayed instead.”
However, it was also stated that “In this incident, had the deteriorating RVR figures been passed to the crew it is unlikely that they would have made a significant change in their approach strategy as the recorded RVR remained above CAT1 limits. The decision to continue the approach was made at decision height with the required visual references and the autopilot was disconnected for a manual landing. Instances of loss of visual references during the landing phase are relatively rare, but not unknown, and occur due to a variety of causes. In the case of this aircraft it could not be determined whether the loss of visual reference was due to a localised area of denser fog, a localised reduction in the quality of runway lighting, or a combination of both.”
It was further notes that “loss of visual references almost at the point of touchdown….could occur due to either changing meteorological conditions or a simple failure of runway lighting. This incident also reinforces the generic advice that crews should remain ‘go-around minded’ throughout the landing phase”
On the subject of the Runway Surface, although it was considered to be of limited consequence to this particular investigation, the level of contamination seen on the surface of Runway 06 and the lack of any evidence that grip testing had been conducted was of concern. The Investigation noted that “the quantity of rubber deposition may reduce the available friction and braking action for landing aircraft on Runway 06, whilst aircraft conducting a rejected takeoff on the reciprocal runway (Runway 24) in wet conditions could suffer a significant loss of braking effectiveness. In the absence of routine grip testing it is unlikely that an airport authority can determine the condition of the runway with regard to either “slippery when wet” or maintenance planning levels.”
The Report concluded that “ The aircraft departed the left side of the runway as a result of the PF’s rudder pedal inputs during the flare which were made during a period when the crew reported that they lost their visual references. In such cases, at this critical phase of flight, it is important that flight crews can recognise and react in a timely manner to unexpected events. The crew recognised the deviation and carried out the initial actions of the go-around within two seconds of the aircraft touching down. This suggests that the decision to go-around had been made before the aircraft actually touched down. The reason for the loss of visual references could not be conclusively proven, but it was considered that local changes in fog density together with variability of runway lighting quality were a factor. The excursion was contained and damage was limited by the timely application of corrective rudder combined with the decision to go-around. However, the aircraft’s left main landing gear did run off the side of the runway for 180 metres.” And that since training for rejected landings is now routinely carried out by UK carriers both during type conversion and recurrent training, no safety recommendations were necessary in that respect
The Five Safety Recommendations made were as follows:
- It is recommended that the Air Traffic Controllers at Nairobi International Airport are provided with appropriate training in the use of the Runway Visual Range measuring equipment which is a function of the Automated Weather Observation System installed at the airport.
- It is recommended that the Kenya Airports Authority review their maintenance programme for runway lighting at Nairobi International Airport to ensure that runway lighting quality complies with ICAO Standards.
- It is recommended that the Kenya Airports Authority take action to ensure that the positioning of the runway edge lights at Nairobi International Airport complies with ICAO Standards.
- It is recommended that the Kenya Airports Authority notify all aircraft operators using Nairobi International Airport of the fact that the runway edge lights are positioned 7.5 m away from the edges of the declared runway surface rather than the maximum of 3 m specified by ICAO.
- It is recommended that the Kenya Airports Authority initiates routine testing to monitor runway friction levels at Nairobi International Airport in order to ensure compliance with the standards required by ICAO.
The Final Report was published in November 2009.