AT76, Jabalpur India, 2022
AT76, Jabalpur India, 2022
On 12 March 2022, an ATR76-600 captain made an unstabilised approach to Jabalpur, India, leading to a first bounce more than halfway along the runway and a final touchdown 400 metres from the runway end. The first officer took control but did not commence a go-around, and the aircraft overran the runway before stopping. The captain had just over four months command experience and had made six similar "high-severity long-flare" approaches in the previous five days. These had gone undetected because although such exceedances were supposedly being tracked by company flight data monitoring, this event was not being tracked.
Description
On 12 March 2022, an ATR76-600 (VT-AIW) operated by Alliance Air on a scheduled domestic passenger flight from Delhi to Jabalpur as 9I-9617 touched down at destination with 400 metres of runway remaining. It exited the runway end before travelling a further 60 metres beyond the end of the runway. The aircraft also deviated 45 metres to the left of its extended centreline before stopping. None of the 59 occupants were injured as a result of the overrun, but the right main gear wheels and hydraulics were damaged as was some of the runway lighting.
Investigation
A Serious Incident investigation was carried out by the Indian Aircraft Accident Investigation Bureau (AAIB). The digital flight data recorder (DFDR) and the two-hour solid-state cockpit voice recorder (SSCVR) were removed from the aircraft, and relevant data from both were downloaded. Recorded Air Traffic Control (ATC) data was also accessed, but although recorded airport closed-circuit CCTV footage should also have been available, the airport’s CCTV recording capability was unserviceable.
The Flight Crew
The 32-year-old captain had a total of 3,000 hours flying experience, of which 2,766 hours had been on type, all obtained since joining Alliance Air as a first officer upon completing her pilot training. She had been released to operate as a first officer eight months after beginning employment and had subsequently been promoted to command four months prior to the investigated event. The 35-year-old first officer had a total of 1,800 hours flying experience, of which 1,476 hours had been on type and had been obtained during just over two years with the operator.
What Happened
The two-hour flight from Delhi was uneventful. Whilst approximately 80 nm from Jabalpur, the latest weather there was obtained which included a surface wind of 310°/05 knots, 5km visibility in haze with no significant cloud and a temperature of 30°C.
When 20 nm from the airport at FL 060, further descent was requested and the flight was cleared for the VOR/DME arc approach to Runway 24. This approach was conducted with the autopilot (AP) engaged in LNAV/VS modes and with selected speed of 107 KIAS. As the aircraft crossed the final approach fix (FAF), clearance to land was given with a spot surface wind of 340°/05 knots. A continuous descent final approach (CDFA) was flown with the landing gear extended and flaps 30 selected with 7 nm to run. The crew became visual with the runway at 1,700 feet agl and, in the presence of thermal updrafts, had set the condition levers to 100% override to maintain the required rate of descent. However, despite the achieved vertical speed averaging 600 fpm between 1,000 feet agl and touchdown, the average speed during that time was VAPP +17 knots.
The runway threshold was crossed at approximately 60 feet. The aircraft then “floated for approximately 19 seconds” before making an initial bounce more than halfway along the 1,988-metres-long dry runway and making a final touchdown at 115 KIAS approximately 400 metres before the runway end. The controller observed this and the absence of a go-around and proactively alerted the airport rescue and firefighting service (RFFS). As the aircraft continued along what remained of the runway, it began to deviate to the left of the runway centreline and continued to do so as it exited the runway at a groundspeed of 56 knots. It eventually stopped approximately 60 metres beyond the end of the runway about 45 metres to the left of the runway extended centreline.
The engines were shut down as the cabin crew confirmed that there were no signs of a hazardous situation. In the absence of any call from the flight deck, as the RFFS vehicles arrived at the scene, the senior cabin crew member (SCCM) called, reporting the situation in the cabin and requested instructions. Once the RFFS had confirmed that there was no fire or fuel leakage, the captain instructed the SCCM to disembark the passengers normally with their hand baggage. This was completed and transport was provided to take the passengers to the terminal building.
An annotated view of the runway showing key power lever positions and speeds. [Reproduced from the Official Report]
Why It Happened
The final approach was continued to a landing despite being well outside the operator’s stabilised approach criteria as specified in Part ‘A’ of the operations manual (OM). There were also no deviation calls from the first officer at any point during the approach or attempted landing and, on taking control without any verbal communication with the captain, he failed to initiate a go-around. Not only was the average speed 17 knots above the 107 knot VAPP during the last 1,000 feet; it rose as high as 134 knots.
Indications that this approach was not an exception to the captain’s performance were found when an examination of the operator’s recent operational flight data monitoring (OFDM) found six instances of a "long flare" in the four days prior to the incident. It was found that although there was a mandatory regulatory requirement for such "long flare" events (and "long landing") events to be defined and tracked by the operator’s OFDM system, this was not happening. It was therefore “considered possible that many such instances had never been monitored." It was considered that this absence of effective flight data monitoring and of corrective actions to address deviations from required standard operating procedures (SOPs) had been a factor in the occurrence, since “no appropriate corrective training or feedback” had been provided to the captain.
The Probable Causes of the Investigated Event were formally recorded as:
- Unstabilised approach
- Inappropriate flare
- Delayed retardation of power levers to flight idle
- Not initiating a go-around when a safe landing could not be assured
One Contributory Factor was also identified as:
- The airline’s inability to monitor deviations from required flight performance in their OFDM and provide corrective training or feedback to pilots.
Safety Action taken by Alliance Air whilst the investigation was in progress was noted as having included a reconfiguration of their OFDM software to capture both "long flare" and "long landing" events.
A total of five Safety Recommendations were issued at the conclusion of the investigation based on its findings as follows:
- that Alliance Air should carry out complete monitoring of flight data for all parameters defined in the flight safety manual and provide feedback and corrective training to its crew based on trends monitored in their OFDM in a timely manner.
- that Alliance Air should review its OFDM parameters to see if it is feasible to include inappropriate flare and delayed retardation of the power levers to flight idle as a parameter to be included in OFDM.
- that Alliance Air should reiterate to its pilots the importance of adherence to stabilised approach criteria.
- that Alliance Air should reiterate to its pilots the importance of proper briefings and callouts for any deviations from the appropriate flight path.
- that the Directorate General of Civil Aviation (DGCA) should issue instructions to all airlines to encourage their pilots to exercise the option of rapid deplaning in emergency situations or incidents when full-scale evacuation may not be required.
The Final Report was completed on 22 December 2022 and published on 20 February 2023.