Description

On 22 May 2011 a Britten-Norman BN2A Islander being operated by Bermudian domiciled carrier Montserrat AW on a scheduled passenger flight from Antigua to Montserrat was considered at risk of an overrun after visual positioning to a day landing on runway 28 at destination in normal ground visibility. The pilot intentionally steered the aircraft off one side of the runway to decrease the degree of potential hazard and the aircraft came to a stop beside the runway and 46 metres from its end without injuries to any of the 8 occupants or damage to the aircraft.

Investigation

An Investigation was carried out by the UK AAIB. It was noted that the pilot was experienced on the aircraft type but that the incident landing was his first on the 596 metre runway 28 at Montserrat which, like the reciprocal direction, had a landing threshold displaced by 28 metres included in the quoted total length. The aircraft was instructed by ATC to join the downwind left hand visual circuit for 05 but after being advised of low cloud approaching the airfield from the west and noting the prevailing light winds, decided to use runway 28 and was so cleared. After touchdown following a light rain shower which had just finished, the aircraft skidded on first contact with the runway and so a go around was flown. The second approach to runway 28 after a low level circuit constrained by the cloud base was again followed by a skid on touchdown although, having been observed by witnesses to land nearer the beginning of the runway than the first time, the landing roll was continued. However, when the pilot became concerned at the poor rate of deceleration, he decided, with about 150 metres of runway remaining, to intentionally exit the paved surface with a turn to the right and subsequently stopped. Witnesses described the runway surface as ‘damp’ rather than ‘wet’.

The Investigation found no faults with the aircraft brakes or braking system and no evidence that it had aquaplaned. A check of the likely aircraft operating weight at the time of the occurrence found that the aircraft had probably been flown at a weight slightly above that calculated by the pilot which was also above the aircraft MLW and both the Antigua and Montserrat WAT limits. This discrepancy was considered to be attributable to an unrealistic assumption about the weight of hold baggage which had been carried on the flight which was assessed likely to have been nearly 4 times greater than that assumed even though only 7 of the 9 available passenger seats had been occupied.

Although figures which enable the calculation of the LDR are not published for weights in excess of the MLW, it was estimated that the LDR at the likely actual landing weight on a dry runway would have been 445 metres with the wind calm and 533 metres with a 5 knot tailwind component compared to the LDA declared in the AIP of 540 metres. It was noted that there are no overrun areas at either end if the runway and both ends are close to steep drops of over 200 feet. It was noted that under the prevailing State Regulations and ICAO Annex14, there was no requirement for the provision of a Runway End Safety Area or a greater obstruction-free area to the side of the runway than existed because of the applicable Airport Category although significant hazards were evident just beyond all of these ‘boundaries’. However, it was also noted that there was a State regulatory requirement for the aerodrome operator to have in place an Safety Management System which included both “processes to identify actual and potential safety hazards and assess the associated risks and processes to develop and implement remedial action necessary to maintain agreed safety performance” which it did not in respect of the risks arising from runway excursions. Further concerns relating to both obstacle clearance and the APAPIs provided to assist approaching aircraft were identified and another recent runway excursion at Montserrat by a different BN2 aircraft also operated by Montserrat AW was noted, although this had been attributed to maintenance error.

The runway surface was found to consist of cambered but un-grooved asphalt. The Investigation found that there was a history of uncertainty in respect of the friction characteristics of the runway for a variety of unrelated reasons.

The airfield was described in the AIP as “VFR Only” with Visual Flight Rules (VFR) weather limits of 5km visibility and a 1500 feet aal cloudbase. At the time of the investigated incident, the aerodrome operator required pilots to undergo a flight check which consisted of six takeoffs and landings at the airport under the supervision of a suitably qualified pilot before being permitted to operate at the airport. There was no written requirement for the use of both runways to be checked and although the airport manager commented that he believed this requirement existed, the incident pilot had only been checked on runway 10.

The Conclusion of the Investigation was as follows:

“No faults were found with the aircraft braking system and there was no evidence that the aircraft had (aquaplaned). An accurate runway friction assessment could not be obtained, but there were no pilot reports of poor friction prior to or after the incident. A tailwind and/or high touchdown airspeed would have increased the landing distance required by the aircraft. Issues identified by the Investigation were pilot training, wind measurements, the aerodrome’s weather limits, the APAPI approach angle, obstructions on the approach and the runway environment.”

During the course of the Investigation, a Special Bulletin (S2-2011) had been published on 21 July 2011 in which three Safety Recommendations were made:

• that the operator of John A Osborne Airport Montserrat should install a windsock and anemometer adjacent to the Runway 28 threshold. [2011-077]

• that the operator of John A Osborne Airport Montserrat in consultation with (their contracted adviser) Air Safety Support International, should revise its operations manual to permit pilots to operate only to and from the runway on which they have been flight checked. [2011-078]

• that the operator of John A Osborne Airport Montserrat should ensure that a runway friction assessment is carried out at the earliest opportunity by a qualified person using suitable equipment [2011-079]

Three further Safety Recommendations were upon the completion of the Investigation:

• that the operator of John A Osborne Airport Montserrat carry out a risk assessment of the hazards associated with runway excursions and implement any necessary mitigating action. [2012-010]

• that the operator of John A Osborne Airport Montserrat remove the obstacles that infringe the ICAO Annex 14 ‘Aerodrome Design and Operations’ takeoff and approach surfaces [2012-011]

• that the operator of John A Osborne Airport Montserrat review the Runway 28 APAPI position and angle setting to improve obstacle clearance on the approach. [2012-012]

The Final Report of the Investigation AAIB Bulletin: 5/2012 EW/C2011/05/04 was published on 10 May 2012.

Further Reading

• Runway Excursion
• Landing on Contaminated Runways
• Beyond the Runway End Safety Area
• Deceleration on the Runway
• Runway Availability
• Beyond the Runway End Safety Area